中英文文献翻译-螺杆压缩机
机械设计制造及自动化中英文对照外文翻译文献
机械设计制造及⾃动化中英⽂对照外⽂翻译⽂献中英⽂对照外⽂翻译⽂献(⽂档含英⽂原⽂和中⽂翻译)使⽤CBN砂轮对螺杆转⼦进⾏精密磨削的⽅法摘要:针对⾼精度加⼯螺杆转⼦,这篇论⽂介绍了利⽤⽴⽅氮化硼(CBN砂轮)对螺杆转⼦进⾏精密磨削的加⼯⽅法。
⾸先,使⽤⼩型电镀CBN砂轮磨削螺杆转⼦。
精确的CBN砂轮轴向轮廓的模型是在齿轮啮合理论的基础上建⽴开发的。
考虑到螺杆转⼦和涂层厚度之间的间隙,主动砂轮的修整引⼊了CBN的砂轮的设计⽅法。
主动砂轮的形状采⽤低速电⽕花线切割技术(低速⾛丝线切割机)进⾏加⼯线CBN主动砂轮的成形车⼑采⽤低速⾛丝机切割机进⾏加⼯。
CBN螺杆转⼦砂轮采⽤本⽂提出的原理进⾏有效性和正确性的验证。
电镀CBN砂轮对螺杆转⼦进⾏加⼯,同时进⾏机械加⼯实验。
在实验中获得的数据达到GB10095-88五级认证。
关键词: CBN砂轮精密磨削螺杆转⼦砂轮外形修整专业术语⽬录:P 螺杆转⼦的参数H 螺杆转⼦的直径Σ砂轮和转⼦的安装⾓度Au 砂轮和转⼦的中⼼距8 螺旋转⼦接触点的旋转⾓x1, y1, z1:转⼦在σ系统中的位置x, y, z: 砂轮端⾯的位置x u ,y u ,z u: x, x y z轴的法向量n x ,ny,nz:X Y Z轴的端⾯法向量n u , nu, nu:砂轮的⾓速度的⽮量:砂轮模块的⾓速度wu:螺旋转⼦的⾓速度w1螺旋转⼦模块的⾓速度转⼦接触点的⾓速度转⼦表⾯接触点的初始速度砂轮表⾯接触点的⾓速度砂轮表⾯接触点的初始速度l砂轮的理论半径砂轮轴的理想位置砂轮表⾯的修改半径砂轮轴的修改位置砂轮表⾯的法向量1.引⾔螺旋转⼦是螺杆压缩机、螺钉、碎纸机以及螺杆泵的关键部分。
转⼦的加⼯精度决定了机械性能。
⼀般来说,铣⼑⽤于加⼯螺旋转⼦。
许多研究者,如肖等⼈[ 1 ]和姚等⼈[ 2 ],对⽤铣⼑加⼯螺旋转⼦做了⼤量的⼯作。
该⽅法可以提⾼加⼯效率。
然⽽,加⼯精度低和表⾯粗糙度不⾼是其主要缺点。
空压机有关结构中英文对照
空气压缩机名词英语解释时间:2010-7—1 11:40:27,点击:252容积式压缩机positive displacement compressor往复式压缩机(活塞式压缩机)reciprocating compressor回转式压缩机rotary compressor滑片式压缩机sliding vane compressor单滑片回转式压缩机single vane rotary compressor滚动转子式压缩机rolling rotor compressor三角转子式压缩机triangle rotor compressor多滑片回转式压缩机multi-vane rotary compressor滑片blade旋转活塞式压缩机rolling piston compressor涡旋式压缩机scroll compressor涡旋盘scroll固定涡旋盘stationary scroll,fixed scroll驱动涡旋盘driven scroll,orbiting scroll斜盘式压缩机(摇盘式压缩机)swash plate compressor斜盘swash plate摇盘wobble plate螺杆式压缩机screw compressor单螺杆压缩机single screw compressor阴转子female rotor阳转子male rotor主转子main rotor闸转子gate rotor无油压缩机oil free compressor膜式压缩机diaphragm compressor活塞式压缩机reciprocating compressor单作用压缩机single acting compressor双作用压缩机double acting compressor双效压缩机dual effect compressor双缸压缩机twin cylinder compressor闭式曲轴箱压缩机closed crankcase compressor开式曲轴箱压缩机open crankcase compressor顺流式压缩机uniflow compressor逆流式压缩机return flow compressor干活塞式压缩机dry piston compressor双级压缩机compound compressor多级压缩机multistage compressor差动活塞式压缩机stepped piston compound compressor, differential piston compressor 串轴式压缩机tandem compressor,dual compressor截止阀line valve, stop valve排气截止阀discharge line valve吸气截止阀suction line valve部分负荷旁通口partial duty port能量调节器energy regulator容量控制滑阀capacity control slide valve容量控制器capacity control消声器muffler联轴节coupling曲轴箱crankcase曲轴箱加热器crankcase heater轴封crankcase seal, shaft seal填料盒stuffing box轴封填料shaft packing机械密封mechanical seal波纹管密封bellows seal转动密封rotary seal迷宫密封labyrinth seal轴承bearing滑动轴承sleeve bearing偏心环eccentric strap滚珠轴承ball bearing滚柱轴承roller bearing滚针轴承needle bearing止推轴承thrust bearing外轴承pedestal bearing臼形轴承footstep bearing轴承箱bearing housing止推盘thrust collar偏心销eccentric pin曲轴平衡块crankshaft counterweight, crankshaft balance weight 曲柄轴crankshaft偏心轴eccentric type crankshaft曲拐轴crank throw type crankshaft连杆connecting rod连杆大头crank pin end连杆小头piston pin end曲轴crankshaft主轴颈main journal曲柄crank arm,crank shaft曲柄销crank pin曲拐crank throw曲拐机构crank-toggle阀盘valve disc阀杆valve stem阀座valve seat阀板valve plate阀盖valve cage阀罩valve cover阀升程限制器valve lift guard阀升程valve lift阀孔valve port吸气口suction inlet压缩机气阀compressor valve吸气阀suction valve排气阀delivery valve圆盘阀disc valve环片阀ring plate valve簧片阀reed valve舌状阀cantilever valve条状阀beam valve提升阀poppet valve菌状阀mushroom valve杯状阀tulip valve缸径cylinder bore余隙容积clearance volume附加余隙(补充余隙)clearance pocket活塞排量swept volume,piston displacement理论排量theoretical displacement实际排量actual displacement实际输气量actual displacement, actual output of gas 气缸工作容积working volume of the cylinder活塞行程容积piston displacement气缸cylinder气缸体cylinder block气缸壁cylinder wall水冷套water cooled jacket气缸盖(气缸头) cylinder head安全盖(假盖)safety head假盖false head活塞环piston ring气环sealing ring刮油环scraper ring油环scrape ring活塞销piston pin活塞piston活塞行程piston stroke吸气行程suction stroke膨胀行程expansion stroke压缩行程compression stroke排气行程discharge stroke升压压缩机booster compressor立式压缩机vertical compressor卧式压缩机horizontal compressor角度式压缩机angular type compressor对称平衡型压缩机symmetrically balanced type compressor空压机英文菜单及报警信息的中英文对照序号英文报警信息中文报警名称1 Emergencey Stop 紧急制动2 Communication Control 通讯控制3 Invalid Access Code 进入码错误4 No fault stored 无故障信息存储5 No fault reset 无复位指示6 No service indicated 无维修指示7 Remote start enable 远程启动8 Remote start enables 远程停机9 Stop machine first 先停机10 Fan motor fault 风扇电机故障11 High air pressure 空气压力过高12 High oil temp fault 油温过高13 Main motor fault 主电机故障14 Pressure probe fault 压力探测器故障15 Rotation fault 转向故障16 Star/delta fault Y/△故障17 Temperature probe fault 温度探测器故障18 Change air filter 更换空气过滤器19 Chang reclaimer element 更换油分离器滤芯20 High oil 油温过高21 Service due 技术服务时间到22 Max overpress 最高过压(自己复位)23 Total 总时数24 Hours on load 负载时数25 Max。
单螺杆型使用的是机械外文文献翻译、中英文翻译、外文翻译
中国地质大学长城学院本科毕业设计外文资料翻译系别:工程技术系专业:机械设计制造及其自动化姓名:李江学号: 052083072012 年 4 月 20 日外文资料翻译译文塑料工业是与国民经济发展和社会文明建设息息相关的重要产业。
塑料工业的机械和装备的水平对该工业的发展起着关键作用。
比起传统的塑料挤出机,单螺杆塑料挤出机有他积极的优势,通常他能加工出高分子量、高粘度热塑性好的塑料。
其中有高生产率低熔点、高熔体强压力的塑料有利于化学降解,产品质量高,品,因此是最佳的塑料生产这使得单螺杆塑料挤出机生产经济,特别适用于稳定的挤压。
螺杆的回转航程和固定筒壁的相互作用是挤出机的泵出过程中必要的参数。
为了运输塑料材料,其摩擦在螺杆的表面要低,但在固定的筒壁要高。
如果达不到这个基本标准,塑料可能会随着螺杆旋转,而不是在轴向/输出方向上移动。
在输出区域,螺杆和机筒的表面通常都覆盖着的溶解物以及来自溶解物和螺杆通道之间的外力,而其除了处理有极高粘性的材料时都是无效的,如硬质PVC 材料和有超高分子量的聚乙烯。
溶解物流在输出部分是受内摩擦系数(粘度)影响,尤其是当模具提供了一个高阻力的熔体流时。
常见且更多使用的单螺杆型使用的是传统设计,即机筒和螺杆保持基本一致的直径,包括具有例如减小螺杆通道体积,有连续可变速度、压力控制,和通风(挥发)系统的挤出机。
一些特殊的设计使用了圆锥或抛物线外形的螺杆,用以达到特殊的混合和捏合效果。
它们可以包含偏心的核心,根据不同坡度变化的动程,揉捏转子,适应性的核心环,和间歇的轴向运动。
桶内可能有螺纹,可伸缩的螺杆形状以及进料设备。
一个成功的挤出操作需要密切注意很多细节,如(1)进给材料的质量和在适当温度下的物质流,(2)足以融化、但不会分解聚合物的温度曲线,以及(3)不会分解塑料的启动和关机。
应采取措施,防止促进塑料表面上水分的胶合和湿气的吸收凝结,如颜料浓缩物中的色素。
表面凝结,可通过储存于密封的塑料容器(吸湿性塑料)中在使用前约24 小时与工作区域保持同温来避免。
螺杆压缩机性能的计算吸入室中占主导地位外文文献翻译、中英文翻译、外文翻译
英文原文3.1 One Dimensional Mathematical Model 51The Conservation of Internal Energyθωω.d dv p Q h m h m dQ du out out in in += (3.1) where θ is angle of rotation of the main rotor, h = h (θ) is specific enthalpy, m ˙ = m ˙ (θ) is mass flow rate p = p (θ), fluid pressure in the working chamber control volume, ˙Q = ˙Q (θ), heat transfer between the fluid and the compressor surrounding, ˙V = ˙V (θ) local volume of the compressor working chamber.In the above equation the subscripts in and out denote the fluid inflow and outflow. The fluid total enthalpy inflow consists of the following components:oil oil g l g l suc suv in in h m h m h m h .,,...m ++= (3.2) where subscripts l , g denote leakage gain suc, suction conditions, and oil denotes oil. The fluid total outflow enthalpy consists of:l l l l dis dis ou out h m h m h m ,,..+= (3.3) where indices l , l denote leakage loss and dis denotes the discharge conditions with m ˙ dis denoting the discharge mass flow rate of the gas contaminated with the oil or other liquid injected.The right hand side of the energy equation consists of the following terms which are model The heat exchange between the fluid and the compressor screw rotors and casing and through them to the surrounding, due to the difference in temperatures of gas and the casing and rotor surfaces is accounted for by the heat transfer coefficient evaluated from the expression Nu = 0.023 Re0.8. For the characteristic length in the Reynolds and Nusselt number the difference between the outer and inner diameters of the main rotor was adopted. This may not be the most appropriate dimension for this purpose, but the characteristic length appears in the expression for the heat transfer coefficient with the exponent of 0.2 and therefore has little influence as long as it remains within the same order of magnitude as other characteristic dimensions of the machine and as long as it characterizes the compressor size. The characteristic velocity for the Re number is computed from the local mass flow and the cross-sectional area. Here the surface over which the heat is exchanged, as well as the wall temperature, depend on the rotation angle θ of the main rotor. The energy gain due to the gas inflow into the working volume is represented by the product of the mass intake and its averaged enthalpy. As such, the energy inflow varies with the rotational angle. During the suction period, gas enters the working volume bringing the averaged gas enthalpy,52 3 Calculation of Screw Compressor Performance which dominates in the suction chamber. However, during the time when the suction port is closed, a certain amount of the compressed gas leaks into the compressor working chamber through the clearances. The mass ofthis gas, as well as its enthalpy are determined on the basis of the gas leakage equations. The working volume is filled with gas due to leakage only when the gas pressure in the space around the working volume is higher, otherwise there is no leakage, or it is in the opposite direction, i.e. from the working chamber towards other plenums.The total inflow enthalpy is further corrected by the amount of enthalpy brought into the working chamber by the injected oil.The energy loss due to the gas outflow from the working volume is defined by the product of the mass outflow and its averaged gas enthalpy. During delivery, this is the compressed gas entering the discharge plenum, while, in the case of expansion due to inappropriate discharge pressure, this is the gas which leaks through the clearances from the working volume into the neighbouring space at a lower pressure. If the pressure in the working chamber is lower than that in the discharge chamber and if the discharge port is open, the flow will be in the reverse direction, i.e. from the discharge plenum into the working chamber. The change of mass has a negative sign and its assumed enthalpy is equal to the averaged gas enthalpy in the pressure chamber.The thermodynamic work supplied to the gas during the compression process is represented by the term pdV d θ . This term is evaluated from the local pressure and local volume change rate. The latter is obtained from the relationships defining the screw kinematics which yield the instantaneous working volume and its change with rotation angle. In fact the term dV/d ϕ can be identified with the instantaneous interlobe area, corrected for the captured and overlapping areas. If oil or other fluid is injected into the working chamber of the compressor, the oil mass inflow and its enthalpy should be included in the inflow terms. In spite of the fact that the oil mass fraction in the mixture is significant, its effect upon the volume flow rate is only marginal because the oil volume fraction is usually very small. The total fluid mass outflow also includes the injected oil, the greater part of which remains mixed with the working fluid. Heat transfer between the gas and oil droplets is described by a first order differential equation. The Mass Continuity Equationout out in in h m h m d m d ...θω= (3.4) The mass inflow rate consists of:oil g l suc in in m m m h m .,..++= (3.5)3.1 One Dimensional Mathematical Model 53The mass outflow rate consists of: .,..l l dis out m m m += (3.6)Each of the mass flow rate satisfies the continuity equationA m ρω=. (3.7) where w [m/s] denotes fluid velocity, ρ – fluid density and A – the flow crosssectionarea. The instantaneous density ρ = ρ(θ) is obtained from the instantaneous mass m trapped in the control volume and the size of the corresponding instantaneous volume V , as ρ = m/V .3.1.2 Suction and Discharge PortsThe cross-section area A is obtained from the compressor geometry and it may be considered as a periodic function of the angle of rotation θ. The suction port area is defined by:⎪⎪⎭⎫ ⎝⎛=suc o suc A A θθπsin ,suc (3.8) where suc means the starting value of θ at the moment of the suction port opening, and A suc , 0 denotes the maximum value of the suction port crosssection area. The reference value of the rotation angle θ is assumed at the suction port closing so that suction ends at θ = 0, if not specified differently.The discharge port area is likewise defined by:⎪⎪⎭⎫ ⎝⎛--=s e c o dis A A θθθθπsin ,dis (3.9)where subscript e denotes the end of discharge, c denotes the end of compression and A dis , 0 stands for the maximum value of the discharge port crosssectional area.Suction and Discharge Port Fluid Velocities)(212h h -=μω (3.10)where μ is the suction/discharge orifice flow coefficient, while subscripts 1 and 2 denote the conditions downstream and upstream of the considered port. The provision supplied in the computer code will calculate for a reverse flow if h 2 < h 1.54 3 Calculation of Screw Compressor Performance3.1.3 Gas LeakagesLeakages in a screw machine amount to a substantial part of the total flow rate and therefore play an important role because they influence the process both by affecting the compressor mass flow rate or compressor delivery, i.e. volumetric efficiency and the thermodynamic efficiency of the compression work. For practical computation of the effects of leakage upon the compressor process, it is convenient to distinguish two types of leakages, according to their direction with regard to the working chamber: gain and loss leakages. The gain leakages come from the discharge plenum and from the neighbouring working chamber which has a higher pressure. The loss leakages leave the chamber towards the suction plenum and to the neighbouring chamber with a lower pressure.Computation of the leakage velocity follows from consideration of the fluid flow through the clearance. The process is essentially adiabatic Fanno-flow. In order to simplify the computation, the flow is is sometimes assumed to be at constant temperature rather than at constant enthalpy. This departure from the prevailing adiabatic conditions has only a marginal influence if the analysis is carried out in differential form, i.e. for the small changes of the rotational angle, as followed in the present model. The present model treats only gas leakage. No attempt is made to account for leakage of a gas-liquid mixture, while the effect of the oil film can be incorporated by an appropriate reduction of the clearance gaps.An idealized clearance gap is assumed to have a rectangular shape and the mass flow of leaking fluid is expressed by the continuity equation:g l l l A m ωρμ=. (3.11)where r and w are density and velocity of the leaking gas, Ag = lg δg the clearance gap cross-sectional area, lg leakage clearance length, sealing line, δg leakage clearance width or gap, μ = μ(Re, Ma) the leakage flow discharge coefficient.Four different sealing lines are distinguished in a screw compressor: the leading tip sealing line formed between the main and gate rotor forward tip and casing, the trailing tip sealing line formed between the main and gate reverse tip and casing, the front sealing line between the discharge rotor front and the housing and the interlobe sealing line between the rotors.All sealing lines have clearance gaps which form leakage areas. Additionally, the tip leakage areas are accompanied by blow-hole areas.According to the type and position of leakage clearances, five different leakages can be identified, namely: losses through the trailing tip sealing and front sealing and gains through the leading and front sealing. The fifth, “throughleakage ” does not directly affect the process in the working chamber, but it passes through it from the discharge plenum towards the suction port. The leaking gas velocity is derived from the momentum equation, which accounts for the fluid-wall friction:3.1 One Dimensional Mathematical Model 5502211=++Dg dxf dpd l ωρωω (3.12)where f (Re, Ma) is the friction coefficient which is dependent on the Reynolds and Mach numbers, Dg is the effective diameter of the clearance gap, Dg ≈ 2δg and dx is the length increment. From the continuity equation and assuming that T ≈ const to eliminate gas density in terms of pressure, the equation can be integrated in terms of pressure from the high pressure side at position 2 to the low pressure side at position 1 of the gap to yield:⎪⎪⎭⎫ ⎝⎛+-==1222122.ln 2m p p a A g l l ςρρωρ (3.13)where ζ = fLg/Dg + Σξ characterizes the leakage flow resistance, with Lg clearance length in the leaking flow direction, f friction factor and ξ local resistance coefficient. ζ can be evaluated for each clearance gap as a function of its dimensions and shape and flow characteristics. a is the speed of sound.The full procedure requires the model to include the friction and drag coefficients in terms of Reynolds and Mach numbers for each type of clearance.Likewise, the working fluid friction losses can also be defined in terms of the local friction factor and fluid velocity related to the tip speed, density, and elementary friction area. At present the model employs the value of ζ in terms of a simple function for each particular compressor type and use. It is determined as an input parameter.These equations are incorporated into the model of the compressor and employed to compute the leakage flow rate for each clearance gap at the local rotation angle θ.3.1.4 Oil or Liquid InjectionInjection of oil or other liquids for lubrication, cooling or sealing purposes, modifies the thermodynamic process in a screw compressor substantially. The following paragraph outlines a procedure for accounting for the effects of oil injection. The same procedure can be applied to treat the injection of any other liquid. Special effects, such as gas or its condensate mixing and dissolving in the injected fluid or vice versa should be accounted for separately if they are expected to affect the process. A procedure for incorporating these phenomena into the model will be outlined later.A convenient parameter to define the injected oil mass flow is the oil-to-gas mass ratio, m oil /m gas, from which the oil inflow through the open oil port, which is assumed to be uniformly distributed, can be evaluated asπ21....z m m m m gas oiloil = (3.14) where the oil-to-gas mass ratio is specified in advance as an input parameter56 3 Calculation of Screw Compressor PerformanceIn addition to lubrication, the major purpose for injecting oil into a compressor is to cool the gas. To enhance the cooling efficiency the oil is atomized into a spray of fine droplets by means of which the contact surface between the gas and the oil is increased. The atomization is performed by using specially designed nozzles or by simple high-pressure injection. The distribution of droplet sizes can be defined in terms of oil-gas mass flow and velocity ratio for a given oil-injection system. Further, the destination of each distinct size of oil droplets can be followed until it hits the rotor or casing wall by solving the dynamic equation for each droplet size in a Lagrangian frame, accounting for inertia gravity, drag, and other forces. The solution of the droplet energy equation in parallel with the momentum equation should yield the amount of heat exchange with the surrounding gas.In the present model, a simpler procedure is adopted in which the heat exchange with the gas is determined from the differential equation for the instantaneous heat transfer between the surrounding gas and an oil droplet. Assuming that the droplets retain a spherical form, with a prescribed Sauter mean droplet diameter dS , the heat exchange between the droplet and the gas can be expressed in terms of a simple cooling law Qo = hoAo (T gas − T oil), where Ao is the droplet surface, Ao = d 2 S π, dS is the Sauter mean diameter of the droplet and ho is the heat transfer coefficient on the droplet surface, determined from an empirical expression. The exchanged heat must balance the rate of change of heat taken or given away by the droplet per unit time, Qo = moc oil dTo/dt = moc oil ωdTo/d θ, where c oil is the oil specific heat and the subscript o denotes oil droplet. The rate of change of oil droplet temperature can now be expressed as:()oilo o gas o c m T T A h d dT ωθ-=00 (3.15) The heat transfer coefficient ho is obtained from:33.06.0Pr Re 6.02u +=N (3.16)Integration of the equation in two time/angle steps yields the new oil droplet temperature at each new time/angle step:k kT T T po gas o +-=1, (3.17)where To,p is the oil droplet temperature at the previous time step and k is the non-dimensional time constant of the droplet, k = τ/Δt = ωτ/Δθ, with τ = moc oil /hoAo being the real time constant of the droplet. For the given Sauter mean diameter, dS , the non-dimensional time constant takes the formθωθω∆=∆=o oil S O o oil o h c d A h c m k 6 (3.18) The derived droplet temperature is further assumed to represent the average temperature of the oil, i.e. T oil ≈ To , which is further used to compute the enthalpy of the gas-oil mixture.3.1 One Dimensional Mathematical Model 57The above approach is based on the assumption that the oil-droplet time constant τ is smaller than the droplet travelling time through the gas before it hits the rotor or casing wall, or reaches the compressor discharge port. This means that heat exchange is completed within the droplet travelling time through the gas during compression. This prerequisite is fulfilled by atomization of the injected oil. This produces sufficiently small droplet sizes to gives a small droplet time constant by choosing an adequate nozzle angle, and, to some extent, the initial oil spray velocity. The droplet trajectory computed independently on the basis of the solution of droplet momentum equation for different droplet mean diameters and initial velocities. Indications are that for most screw compressors currently in use, except, perhaps for the smallest ones, with typical tip speeds of between 20 and 50m/s, this condition is well satisfied for oil droplets with diameters below 50 μm. For more details refer to Stosic et al., 1992.Because the inclusion of a complete model of droplet dynamics would complicate the computer code and the outcome would always be dependant on the design and angle of the oil injection nozzle, the present computation code uses the above described simplified approach. This was found to be fully satisfactory for a range of different compressors. The input parameter is only the mean Sauter diameter of the oil droplets, dS and the oil properties – density, viscosity and specific heat.3.1.5 Computation of Fluid PropertiesIn an ideal gas, the internal thermal energy of the gas-oil mixture is given by:()()()()oil oil gasoil gas mcT pV mcT mRT mu mu T +-=+-=+=11γγ (3.19)where R is the gas constant and γ is adiabatic exponentHence, the pressure or temperature of the fluid in the compressor working chamber can be explicitly calculated by input of the equation for the oil temperature T oil:()()()()()oilOIL mc mR k mcT U k T ++-+-=111γ (3.20) If k tends 0, i.e. for high heat transfer coefficients or small oil droplet size, the oil temperature fast approaches the gas temperature.In the case of a real gas the situation is more complex, because the temperature and pressure can not be calculated explicitly. However, since the internal energy can be expressed as a function of the temperature and specific volume only, the calculation procedure can be simplified by employing the internal energy as a dependent variable instead of enthalpy, as often is the practice. The equation of state p = f 1(T,V ) and the equation for specific internal energy u = f 2(T,V ) are usually decoupled. Hence, the temperature can be calculated from the known specific internal energy and the specific volume obtained from the solution of differential equations, whereas the pressure中文译文33.1一维数学模型 51内部能量守恒θωω.d dv p Q h m h m dQ du out out in in += (3.1) 其中θ是角度的旋转的主旋翼h =h ( θ )的比焓,m ˙ =m ˙ ( θ )是质量流率p = ( θ ) ,工作腔的控制体积中的流体压力, ˙ Q = Q˙( θ )的流体之间的热传递和压缩机周围, ˙ V = ˙V ( θ ) ,压缩机工作腔中的本地卷。
阀门术语中英文对照
容积式压缩机positive displacement compressor往复式压缩机(活塞式压缩机)reciprocating compressor回转式压缩机rotary compressor滑片式压缩机sliding vane compressor单滑片回转式压缩机single vane rotary compressor滚动转子式压缩机rolling rotor compressor三角转子式压缩机triangle rotor compressor多滑片回转式压缩机multi-vane rotary compressor滑片blade旋转活塞式压缩机rolling piston compressor涡旋式压缩机scroll compressor涡旋盘scroll固定涡旋盘stationary scroll, fixed scroll驱动涡旋盘driven scroll, orbiting scroll斜盘式压缩机(摇盘式压缩机)swash plate compressor斜盘swash plate摇盘wobble plate螺杆式压缩机screw compressor单螺杆压缩机single screw compressor阴转子female rotor阳转子male rotor主转子main rotor闸转子gate rotor无油压缩机oil free compressor膜式压缩机diaphragm compressor活塞式压缩机reciprocating compressor单作用压缩机single acting compressor双作用压缩机double acting compressor双效压缩机dual effect compressor双缸压缩机twin cylinder compressor闭式曲轴箱压缩机closed crankcase compressor开式曲轴箱压缩机open crankcase compressor顺流式压缩机uniflow compressor逆流式压缩机return flow compressor干活塞式压缩机dry piston compressor双级压缩机compound compressor多级压缩机multistage compressor差动活塞式压缩机stepped piston compound compressor, differential piston compressor 串轴式压缩机tandem compressor, dual compressor截止阀line valve, stop valve排气截止阀discharge line valve吸气截止阀suction line valve部分负荷旁通口partial duty port能量调节器energy regulator容量控制滑阀capacity control slide valve容量控制器capacity control消声器muffler联轴节coupling曲轴箱crankcase曲轴箱加热器crankcase heater轴封crankcase seal, shaft seal填料盒stuffing box轴封填料shaft packing机械密封mechanical seal波纹管密封bellows seal转动密封rotary seal迷宫密封labyrinth seal轴承bearing滑动轴承sleeve bearing偏心环eccentric strap滚珠轴承ball bearing滚柱轴承roller bearing滚针轴承needle bearing止推轴承thrust bearing外轴承pedestal bearing臼形轴承footstep bearing轴承箱bearing housing止推盘thrust collar偏心销eccentric pin曲轴平衡块crankshaft counterweight, crankshaft balance weight 曲柄轴crankshaft偏心轴eccentric type crankshaft曲拐轴crank throw type crankshaft连杆connecting rod连杆大头crank pin end连杆小头piston pin end曲轴crankshaft主轴颈main journal曲柄crank arm, crank shaft曲柄销crank pin曲拐crank throw曲拐机构crank-toggle阀盘valve disc阀杆valve stem阀座valve seat阀板valve plate阀盖valve cage阀罩valve cover阀升程限制器valve lift guard阀升程valve lift阀孔valve port吸气口suction inlet压缩机气阀compressor valve吸气阀suction valve排气阀delivery valve圆盘阀disc valve环片阀ring plate valve簧片阀reed valve舌状阀cantilever valve条状阀beam valve提升阀poppet valve菌状阀mushroom valve杯状阀tulip valve缸径cylinder bore余隙容积clearance volume附加余隙(补充余隙)clearance pocket活塞排量swept volume, piston displacement理论排量theoretical displacement实际排量actual displacement实际输气量actual displacement, actual output of gas 气缸工作容积working volume of the cylinder活塞行程容积piston displacement气缸cylinder气缸体cylinder block气缸壁cylinder wall水冷套water cooled jacket气缸盖(气缸头)cylinder head安全盖(假盖)safety head假盖false head活塞环piston ring气环sealing ring刮油环scraper ring油环scrape ring活塞销piston pin活塞piston活塞行程piston stroke吸气行程suction stroke膨胀行程expansion stroke压缩行程compression stroke排气行程discharge stroke升压压缩机booster compressor立式压缩机vertical compressor卧式压缩机horizontal compressor角度式压缩机angular type compressor对称平衡型压缩机symmetrically balanced type compressor CQ螺纹球阀CQ Thread Ball ValvesL形三通式L-pattern three wayT形三通式T-pattern three way安全阀Safety valve暗杆闸阀Inside screw nonrising stem type gate valve百叶窗;闸板shutter百叶窗式挡板louver damper摆阀式活塞泵swing gate piston pump保温式Steam jacket type报警阀alarm valve报警阀;信号阀;脉冲阀sentinel valve背压调节阀back pressure regulating valve背压率Rate of back pressuec本体阀杆密封body stem seal波纹管阀Bellows valves波纹管密封阀bellow sealed valve波纹管密封式Bellows seal type波纹管平衡式安全阀Bellows seal balance safety valve波纹管式减压阀Bellows reducing valve波纹管式减压阀Bellows weal reducing valve薄膜thin film薄膜;隔膜diaphragm薄膜式减压阀Diapjragm reducing valve薄型闸阀Thin Gate Valves不封闭式Unseal type槽车球阀Tank Lorry Ball Valves颤振Flutter常闭式Normally closed type常开式Normally open type超低温阀门Cryogenic valve超高压阀门Super high pressure valve超过压力Overpressure of a safety valve衬胶隔膜阀rubber lined diaphragm衬胶截止阀rubber lined globe valve垂直板式蝶阀Vertical disc type butterfly valve磁耦合截止阀Magnetic Co-operate Globe Valves带补充载荷的安全阀Supplementary loaded safety valve带辅助装置的安全阀Assisted safety valve单阀碟双面平行密封闸阀parallel single disk gate valve单口排气阀Single Opening Exhaust Valves单向阀Non-return Valve单闸板Single gate disc单闸板平板闸阀Single Disc Flat Gate Valves弹簧薄膜式减压阀Spring diaphragm reducing valve弹簧式安全阀Direct spring loaded safety valve弹簧座Spring plate弹性闸板Flexible gate disc当量计算排量Equivalent calculated capacity挡板damper导阀Pilot valve导向套Valve guide disc guide低温阀门Sub-zero valve低压阀门Low pressure valve底阀bottom valve底阀Foot valve电磁动装置Eletro magnetic actuator电磁阀magnetic valve电磁阀solenoid valve电磁-液动装置Eletro magnetichydraulic actuator电动阀mortor operated valve电动阀motorized valve电动截止阀Electric Actuated Stop Valves电动平行式双闸板闸板Electric Double Disk Parallel Gate Valves 电动楔式闸阀Electric Actuated Wedge Gate Valves电动装置Electric actuator电-液动装置Eletro hydraulic actuator电液伺服阀electro-hydraulic servovalve调节弹簧Regulation spring调节阀adjusting valve调节阀control valve调节阀regulating valve调节螺套Adjusting bolt Adjusting screw调节圈Adjusting ring蝶板Disc蝶阀;瓣阀butterfly valve蝶阀;瓣阀;拍门;铰链阀flap valve蝶式缓冲止回阀Butterfly Type Non-slam Check蝶式止回阀Butterfly swing check valve定比减压阀Proprutioning pressure reducing valve定差减压阀Fixed differential reducing valve定值减压阀Fixed pressure reducing valve动态特性Dynamic characteristics对焊连接阀Buttwelding valves对夹蝶板阀Wafer plate valves对夹式衬胶蝶阀Wafer Type Butterfly Valves with Rubber Itning对夹式阀门Clamp valves对夹式止回阀Wafer Check Valves额定排量Certified capacity额定排量系数Derated coefficient of discharge 二通阀Two-way valves阀valve阀板valve deck plate阀板valve plate阀板式活塞泵valve deck plate type piston pump 阀板式活塞泵valve plate type piston pump阀瓣Disc阀操纵杆valve operating rod阀痤槽valve seat recess阀挡valve grid阀挡valve positioner阀挡valve stop阀导杆valve tail rod阀导向器valve guide阀盖bonnet阀盖衬套bonnet bush阀盖垫片bonnet gasket阀杆stem阀杆valve rod阀杆valve spindle阀杆端部尺寸Dimmension of valve stem end阀杆环stem ring阀杆螺母Yoke bushing Yoke nut阀杆填料stem packing阀杆头部尺寸Dimension of valve stem head阀簧valve spring阀簧压板valve spring plate阀控水锤泵valve-controlled hydraulic ram阀框架valve yoke阀门Valve阀门传动装置valve bandle set阀门和管件Valves and Fittings阀门盘根valve packing阀门手柄valve handle阀盘disc阀盘valve disc阀片Disc阀球valve ball阀驱动臂valve driving arm阀驱动臂valve motion arm阀式活塞valve type piston阀式活塞valve type bucket阀室式活塞泵valve box type piston pump阀室式活塞泵(美)valve pot type piston pump阀抬起装置valve lifting device阀体body阀体valve body阀箱valve box阀箱valve cage阀箱valve chest阀箱;阀限位器valve guard阀箱盖cover for valve box阀箱盖valve box cover阀箱式活塞泵(美)turret type piston pump阀形活塞泵valve type piston pump阀座Seat ring阀座valve carrier阀座valve seat(body seat)阀座;阀盘valve seat阀座环seat ring阀座密封嵌条sealing strip for valve seat法兰flange法兰堵头blind flange法兰端flange end法兰接头flange joint法兰连接紧固件(双头螺栓和螺帽)flange bolting 法兰密封面,法兰面flange facing法兰面加工flange facing finish法兰球阀Flange Ball Valves翻板阀Flap反冲盘Disc holder反向作用式减压阀Reverse acting reducing valve反向作用式减压阀Reverse acting reducing valve放空阀emptying valve放气阀air vent valve;vent valve放气阀;排气阀air evacuation valve放泄阀escape valve分置阀室式活塞泵separate valve box type piston pump 分置阀室式活塞泵(美)side pot type piston pump封闭式Seal type浮动式球阀Float ball valve浮球Ball float浮球阀Float Valve阀门种类中英文对照中文英文中文英文阀门valve 通用阀门General valve球阀Ball valve 槽车球阀Tank Lorry Ball ValvesCQ 螺纹球阀CQ Thread Ball Valves 法兰球阀Flange Ball Valves衬里球阀Lining Ball Valves API球阀API Ball valves浮动球阀Float Ball valves 固定球阀Trunnion-mounted Ball valves三通法兰球阀Three Ball valves V 型调节球阀V-Regulating Ball valves闸阀Gate valve 楔式闸阀Wedge Gate Valves法兰闸阀Flange Gate Valves 排渣闸阀Scum Gate V alves薄型闸阀Thin Gate Valves 水封闸阀Water Seal Gate Valves电动楔式闸阀Electric Actuated Wedge Gate Valves 平板闸阀Flat Gate Valve单闸板平板闸阀Single Disc Flat Gate Valves 电动平行式双闸板闸阀Electric Double Disk Parallel Gate Valves双闸板平板闸阀Double Disc Flat Gate Valves 明杆平行式双闸板闸阀Double Disk Parallel Gate Valves截止阀Globe valve 角式截止阀Angle Globe valves门角式截止阀Angle Type Globe V alves 电动截止阀Electric Actuated Stop V alves法兰截止阀Flange Globe Valves 衬里截止阀Lining Globe V alves直流式截止阀Oblique Stop Valves 柱塞截止阀Plunger Globe Valves止回阀Check valve 旋启式止回阀Swing Check Valves对夹式止回阀Wafer Check Valves 蝶式缓冲止回阀Butterfly Type Non-slam Check蝶式缓冲止回阀Butterfly Type Non-slam Check 升降式止回阀Lift Check Valves衬里止回阀Lining Check Valves 微阻缓闭止回阀Tiny Drag Slow Shut Check V alves立式止回阀Vertical Lift Check Valves蝶阀Butterfly Valve 对夹蝶板阀Wafer plate valve衬里蝶阀Lining Butterfly Valves 蜗轮传动蝶阀Butterfly Valves with Gear Actuator对夹式衬胶蝶阀Wafer Type Butterfly Valves with Rubber Itning 金属密封蝶阀Hard Seal Butterfly Valves隔膜阀diaphragm valve 旋塞阀plug valve安全阀safety valve 管道安全阀Piping Safety Valves回转阀rotary valve 减压阀pressure reducing valve减速阀Deceleration valves 泄压阀Decompression valves蒸汽疏水阀Automatic steam trap valve 组合阀Combination valves针形阀Pintle valve 仪表阀Gauge Valves仪表针形截止阀Meter Needle Type Globe Valves 空气阀门Air valves排灰阀Ash valves 吸(抽)气阀Aspirating valves辅助(副)阀Auxiliary valves 平衡阀Balance valves波纹管阀Bellows valves 泄料(放空,排污)阀Blowdown valves制动阀Brake valves 对焊连接阀Buttwelding valves地下管道阀Culvert valves 双口排气球Double Opening Exhaust Valves排水阀Drainage valves 紧急切断阀Emergeny Cut-off Valves排气阀Exhaust valves 浮球式疏水阀Free Float Type Steam Trap手动阀Hand-operated valves 液压继动阀Hydraulic relay valves限位阀Limit valves 衬里三通旋塞阀Lining T-Cock Valves 液位计Liquid Indicator 浆液阀Parallel Slide Valves柱塞阀Plunger valves 压力(増压)阀Pressure valve快速排污阀Quick Draining Valves 电磁阀Solenoid valves 过滤器Strainer 节流阀Throttle Valves阀门术语中英文对照中文英文口径bore公称通径Nominal diameter公称压力Nominal pressure工作温度Operating temperature工作压差Operating pressure工作背压Operating back pressure关闭压力Lockup pressure开启高度Lift壳体试验Shell test壳体试验压力Seal test pressure连接尺寸Conncetion cimension连接形式Type of connection流道面积Flow area流道直径Flow diameter流量孔板flow orifice plate流量特性Flow characteristics流量特性偏差Flow characteristics derivation漏汽量Steam loss密封面Sealing face密封试验Seal test密封试验压力Seal test pressure上密封Back seat上密封试验Back seal test适用介质Suitable medium适用温度Suitable temperature最大过冷度Maximum subcoold temperature最大流量Maximum flow rate最大压差Maximum differential pressure最低工作压力Minimum operating pressure最高背压率Maximum rate of back pressure最高允许压力Maximum allowable pressure最小过冷度Minimum subcooled temperature最小压差Minimum differntial pressure常闭式Normally closed type常开式Normally open type阀门零配件中英文对照中文英文阀体body阀盖bonnet阀盖垫片bonnet gasket阀盖衬套bonnet bush阀瓣Disc阀箱valve box阀座Seat ring阀杆Stem阀杆螺母Yoke bushing Yoke nut 法兰flange填料Packing填料垫Packing seat密封件Sealing挡板damper导向套Valve guide disc guide弹簧座Spring plate轴套Axis Guide球、球芯Ball密封圈Ball seat螺母Mut螺栓Screw弹簧Spring闸板Wedge Disc。
螺杆压缩机机械外文文献翻译、中英文翻译、外文翻译
英文原文Screw CompressorThe Symmetric profile has a huge blow-hole area which excludes it from any compressor applicat -ion where a high or even moderate pressure ratio is involved. However, the symmetric profile per -forms surprisingly well in low pressure compressor applications.More details about the circular p -rofile can be found in Margolis, 1978.2.4.8 SRM “A” ProfileThe SRM “A” profile is shown in Fig. 2.11. It retains all the favourable features of the symmetric profile like its simplicity while avoiding its main disadvantage,namely, the large blow-hole area. The main goal of reducing the blow hole area was achieved by allowing the tip points of the main and gate rotors to generate their counterparts, trochoids on the gate and main rotor respectively. T -he “A” profile consists mainly of circles on the gate rotor and one line which passes through the gate rotor axis.The set of primary curves consists of: D2C2, which is a circle on the gate rotor with the centre on the gate pitch circle, and C2B2, which is a circle on the gate rotor, the centre of whi ch lies outside the pitch circle region.This was a new feature which imposed some problems in the generation of its main rotor counterpart, because the mathematics used for profile generation at tha -t time was insufficient for general gearing. This eccentricity ensured that the pressure angles on th -e rotor pitches differ from zero, resulting in its ease of manufacture. Segment BA is a circle on th -e gate rotor with its centre on the pitch circle. The flat lobe sides on the main and gate rotors weregenerated as epi/hypocycloids by points G on the gate and H on the main rotor respectively. GF2 is a radial line at the gate rotor. This brought the same benefits to manufacturing as the previously mentioned circle eccentricity onFig. 2.11 SRM “A” Profile2.4 Review of Most Popular Rotor Profiles 31 the opposite lobe side. F2E2 is a circle with the cent -re on the gate pitch and finally, E2D2 is a circle with the centre on the gate axis.More details on t -he “A” profile are published by Amosov et al., 1977 and by Rinder, 1979.The “A” profile is a go od example of how a good and simple idea evolved into a complicated result. Thus the “A” pro file was continuously subjected to changes which resulted in the “C” profile. This was mainly gen erated to improve the profile manufacturability. Finally, a completely new profile, the“D” profile was generated to introduce a new development in profile gearing and to increase the gate rotor tor -que.Despite the complexity o f its final form the “A” profile emerged to be the most popular scre -w compressor profile, especially after its patent expired.2.4.9 SRM “D” ProfileThe SRM “D” profile, shown in Fig. 2.12, is generated exclusively by circles with the centres off the rotor pitch circles.Similar to the Demonstrator, C2D2 is an eccentric circle of radius r3 onthe gate rotor. B1C1 is an eccentric circle of radius r1, which, together withthe small circular arc A1J1 of radius r2, is positioned on the main rotor. G2H2is a small circular arc on the gate rotor and E2F2 is a circular arc on the gaterotor. F2G2 is a relatively large circular arc on the gate rotor which produces a corresponding curve of the smallest possible curvature on the main rotor.Both circular arc, B2C2 and F2G2 ensure a large radius of curvature in the pitch circle area. This avoids high stresses in the rotor contact region.Fig. 2.12 SRM “D” ProfileThe “G” profile was introduced by SRM in the late nineteen nineties as a replacement for the “D” rotor and is shown in Fig. 2.13. Compared with the“D” rotor, the “G” rotor has the unique feature of two additional circles in the addendum area on both lobes of the main rotor, close to the pitch circle.This feature improves the rotor contact and, additionally, generates shorter sealing lines. This can be seen in Fig. 2.13, where a rotor featuring “G” profile characteristics only on its flat side through segment H1I1 is presented.Fig. 2.13 SRM “G” Profile2.4.11 City “N” Rack Generated Rotor Profile“N” rotors are calculated by a rack generation procedure. This distinguishes them from any others. In this case, the large blow-hole area, which is a characteristic of rack generated rotors, is overcome by generating the high pressure side of the rack by means of a rotor conjugate procedure. This undercuts the single appropriate curve on the rack. Such a rack is then used for profiling both the main and the gate rotors. The method and its extensions were used by the authors to create a number of different rotor profiles, some of them used by Stosic et al., 1986, and Hanjalic and Stosic, 1994. One of the applications of the rack generation procedure is described in Stosic, 1996.The following is a brief description of a rack generated “N” rotor profile,typical of a family of rotor profiles designed for the efficient compression of air,common refrigerants and a number of process gases. The rotors are generated by the combined rack-rotor generation procedure whose features are such that it may be readily modified further to optimize performance for any specific application.2.4 Review of Most Popular Rotor Profiles 33The coordinates of all primary arcs on the rack are summarized here relative to the rack coordinate system. The lobe of the rack is divided into several arcs. The divisions between the profile arcs are denoted by capital letters and each arc is defined separately, as shown in the Figs.2.14 and 2.15 where the rack and the rotors are shown.Fig. 2.14 Rack generated “N” ProfileFig. 2.15 “N” rotor primary curves g iven on rack34 2 Screw Compressor GeometryAll curves are given as a “general arc” expressed as: axp + byq = 1. Thus straight lines, circles, parabolae, ellipses and hyperbolae are all easily described by selecting appropriate values for parameters a, b, p and q.Segment DE is a straight line on the rack, EF is a circular arc of radius r4,segment FG is a straight line for the upper involute, p = q = 1, while segment GH on the rack is a meshing curve generated by the circular arc G2H2 on the gate rotor. Segment HJ on the rack is a meshing curve generated by the circular arc H1J1 of radius r2 on the main rotor. Segment JA is a circular arc of radius r on the rack, AB is an arc which can be either a circle or a parabola, a hyperbola or an ellipse, segment BC is a straight line on the rack matching the involute on the rotor round lobe and CD is a circular arc on the rack, radius r3.More details of the “N” profile can be found in Stosic, 1994.2.4.12 Characteristics of “N” ProfileSample illustrations of the “N” profile in 2-3, 3-5, 4-5, 4-6, 5-6, 5-7 and 6-7 configurations are given in Figs. 2.16 to Fig. 2.23. It should be noted that all rotors considered were obtained automatically from a computer code by simply specifying the number of lobes in the main and gate rotors, and the lobe curves in the general form.A variety of modified profiles is possible. The “N” profile design is a compromise between full tightness, small blow-hole area, large displacement.Fig. 2.16 “N” Rotors in 2-3 configurationFig. 2.17 “N” Rotors in 3-5 configurationFig. 2.18 “N” Rotors in 4-5 configurationFig. 2.19 “N” Rotors in 4-6 configurationFig. 2.20 “N” Rotors compared with “Sigma”, SRM “D” and “Cyclon” rotorsFig. 2.21 “N” Rotors in 5-6 configurationFig. 2.22 “N” Rotors in 5-7 configurationFig. 2.23 “N” rotors in 6/7 configurationsealing lines, small confined volumes, involute rotor contact and proper gate rotor torque distribution together with high rotor mechanical rigidity.The number of lobes required varies according to the designated compressor duty. The 3/5 arrangement is most suited for dry air compression, the 4/5 and 5/6 for oil flooded compressors with a moderate pressure difference and the 6/7 for high pressure and large built-in volume ratio refrigeration applications.Although the full evaluation of a rotor profile requires more than just a geometric assessment, some of the key features of the “N” profile may be readily appreciated by comparing it with three of the most popular screw rotor profiles already described here, (a) The “Sigma” profile by Bammert,1979, (b) the SRM “D” profile by Astberg 1982, and (c) the “Cyclon” profile by Hough and Morris, 1984. All these rotors are shown in Fig. 2.20 where it can be seen that the “N” profiles have a grea ter throughput and a stiffer gate rotor for all cases when other characteristics such as the blow-hole area, confined volume and high pressure sealing line lengths are identical.Also, the low pressure sealing lines are shorter, but this is less important because the corresponding clearance can be kept small.The blow-hole area may be controlled by adjustment of the tip radii on both the main and gate rotors and also by making the gate outer diameter equal to or less than the pitch diameter. Also the sealing lines can be kept very short by constructing most of the rotor profile from circles whose centres are close to the pitch circle. But, any decrease in the blow-hole area will increasethe length of the sealing line on the flat rotor side. A compromise betweenthese trends is therefore required to obtain the best result.2.4 Review of Most Popular Rotor Profiles 39Rotor instability is often caused by the torque distribution in the gate rotor changing direction during a complete cycle. The profile generation procedure described in this paper makes itpossible to control the torque on the gate rotor and thus avoid such effects. Furthermore, full involute contact between the “N” rotors enables any additional contact load to be absorbed more easily than with any other type of rotor. Two rotor pairs are shown in Fig. 2.24 the first exhibits what is described as “negative” gate rotor torque while the second shows the more usual “positive” torque.Fig. 2.24 “N” with negative torque, left and positive torque, right2.4.13 Blower Rotor ProfileThe blower profile, shown in Fig. 2.25 is symmetrical. Therefore only one quarter of it needs to be specified in order to define the whole rotor. It consists of two segments, a very small circle on the rotor lobe tip and a straight line. The circle slides and generates cycloids, while the straight line generates involutes.Fig. 2.25 Blower profile中文译文螺杆压缩机螺杆压缩机的几何形状对称分布有一个巨大的吹孔面积不包括它任何压缩机应用在高或中等压力比参与。
喷油螺杆压缩机的流量分析外文文献翻译、中英文翻译、外文翻译
中文译文4.3 在喷油螺杆压缩机的流量4.3.1 网格生成的油润滑压缩机阳极和阴极的转子有40个数值细胞沿各叶片间的圆周方向,6细胞在径向和轴向方向上的112。
这些形式为转子和壳体444830细胞总数。
为了避免需要增加网格点的数量,如果一个更精确的计算是必需的,一个适应的方法已应用于边界的定义。
时间变化的数量为25,在这种情况下,一个内部循环。
的对阳极的转子转一圈所需的时间步骤的总数是那么125。
在转子中的细胞数为每个时间步长保持相同。
以实现这一目标,一个特殊的网格移动程序开发中的时间通过压缩机转速的确定步骤,正如4章解释。
对于初始时间步长的数值网格图4-15提出。
图4数值网格喷油螺杆压缩机444830细胞4.3.2数学模型的油润滑压缩机数学模型的动量,能量,质量和空间方程问题,如第2.2节所描述的,但一个额外的方程的标量属性油的浓度的增加使石油对整个压缩机性能的影响进行计算。
本构关系是一样的前面的例子。
石油是一种被动的物种在模型处理,这不混合液体-空气的背景。
对空气的影响占通过物质和能量的来源是加上或减去的主要流模型相应的方程。
在这种情况下,动量方程通过拖曳力的影响如前所述。
建立工作条件和从吸气开始全方位1巴压力获得6,7压力的增加,8和9条近450000细胞放电,数值网格对于每一种情况下只有25时间步骤来获得所需的工作条件,其次是进一步的25的时间的步骤来完成一个完整的压缩机循环。
每个时间步所需的约30分钟的运行时间在一个800 MHz的AMD 速龙处理器计算机内存需要约450 MB。
4.3.3对油的数值模拟和实验结果的比较—淹没式压缩机在压缩机中的腔室,在压缩机内的循环的实验得到的压力历史和测得的空气流量和压缩机功率的情况下,测量的速度场担任了宝贵的基础,以验证CFD计算的结果。
要获得这些值,5/6喷油压缩机中,已经描述的,测试安装在压缩机实验室在城市大学伦敦,如图4-16上的钻机。
4-16喷油螺杆空气压缩机5 / 6-128mm(= 90mm)在测试床4.3流的喷油螺杆压缩机该试验台满足螺杆压缩机的接受所有pneurop /程序的要求试验。
压榨机文献翻译
单螺杆挤压数字模拟Raman V. Chiruvella Y. Jaluria” & Mukund V. Karwe罗格斯大学机械和航空航天工程学系美国,新泽西州08903,新不伦瑞克罗格斯大学食品科学系收稿于1904年7月29日,于修订1995年10月10日;接受于2月9日1996摘要我们提出了一种数字化研究淀粉基材料在单螺杆挤出机中的挤压制作。
低水分级和高温通常被认为是实际中会遇到的典型现象。
研究淀粉在实验条件下的转化过程,往往要研究的是最后几转,其中实验材料被当做是一种非牛顿流体。
基于有限差分法的数值模拟方法近似于求解非线性方程,能量和质量守恒的非牛顿流体发生物理和化学变化。
这个问题的初始条件是从实验观察中得来的。
螺杆的参数配置及操作参数,例如机筒温度、螺杆转速、生产能力等,通过变换这些参数来测试它们对淀粉转换的影响。
研究发现,28%的转换由于粘性耗散单独获得,而61%转换往往通过机筒温度提高25°C以上。
同时也观察到在任何螺杆转速较小流量较小的卡死会引起直径发生更大程度的转换。
此外,还发现螺杆压缩比的上压力会显著影响体积温度和平均停留时间的上升。
随着压缩比的提高温度增高但停留时间的减小。
前者的效果增加会使后者转化程度减小。
因此,在最低程度的转换存在一个压缩比。
版权归农业科学。
1996。
作者可做对应处理。
符号注释A M Amioca重量的百分比(干固体)A W 水在Amioca中的重量百分比B 螺杆通道的宽度C 水分含量百分比(机)C P 定压比热容Db 桶直径Es 剪切活化能E T 热活化能H 英吉利螺杆的入口Hz 螺旋通道离z距离L 螺杆的轴向长度Lr 总轴向长度的流变区K 反应动力学速率Km 含水率,载荷系数K T热效果反应动力学速率Ks 应动力学速率由于剪切效应M 秩序的反应M 质量流率,或者吞吐量(公斤/ 秒)m 质量流率,或者吞吐量(公斤/ h)n 在幂律指数载荷N 螺杆转速(rpm)P 局部承压R 通用气体常数(calimole),载荷(20)S 焓变化联系在一起的糊化或转换(W / m”) t 时间(s)T 局部温度u 在x方向的速度Va 速度的螺旋轴的方向Vbx x方向桶速度Vbz z方向桶速度w 速度在z方向移动x 距离出发进行跨海峡协调X 程度的转换在百分比y 协调距离垂直于螺杆及桶的表面z 协调的距离,下行信道方向(沿螺杆螺旋)下行信道中的流变区距离zir希腊字母β锥形螺杆压缩比(H/H, = Zmax)γ剪切速率Κ导热系数的食品物料被挤压μ粘度ρ密度τ(平均剪切应力/平方厘米)载荷(21)θ螺旋角度δ螺旋轴平行的方向挤压成型工艺过程的数学模拟下标a 轴向b 桶dc 下行道dev 成熟介绍挤压烹饪是用作制造食物产品,如入口早餐麦片,扩大了点心、面食、扁平的面包,汤和饮料基地等。
螺杆压缩机Screw compressors_ECDP
2///2008-05-30 16:57
TYPE OF SCREW COMPRESSOR
OIL FREE SCREW COMPRESSOR
2///2008-05-30 16:57
TYPE OF SCREW COMPRESSOR
OIL FREE SCREW COMPRESSOR
The following are the major characteristics of the oil-free screw compressors:
Positive Displacement Type
Dynamic Type Positive Displacement type compressors are those in which successive volumes of gas are confined within some type of enclosure (compression chamber) and elevated to a higher pressure. Eg. Reciprocating compressors, screw
2///2008-05-30 16:57
INTRODUCTION
COMPRESSOR :Any machine which uses external energy to increase the pressure of a gas/ mixture of gases is called a compressor. Compressors are broadly classified into the following two categories-
2///2008-05-30 16:57
BASIC OF SCREW COMPRESSORS
包络法的资产负债-螺杆压缩机转子外文文献翻译、中英文翻译、外文翻译
英文原文AEnvelope Method of GearingFollowing Stosic 1998, screw compressor rotors are treated here as helical gears with nonparallel and nonintersecting, or crossed axes as presented at Fig. A.1. x01, y01 and x02, y02are the point coordinates at the end rotor section in the coordinate systems fixed to the main and gate rotors, as is presented in Fig. 1.3. Σ is the rotation angle around the X axes. Rotation of the rotor shaft is the natural rotor movement in its bearings. While the main rotor rotates t hrough angle θ, the gate rotor rotates through angle τ = r1w/r2wθ = z2/z1θ, where r w and z are the pitch circle radii and number of rotor lobes respectively. In addition we define external and internal rotor radii: r1e= r1w+ r1 and r1i= r1w− r0. The dista nce between the rotor axes is C = r1w+ r2w. p is the rotor lead given for unit rotor rotation angle. Indices 1 and 2 relate to the main and gate rotor respectively.Fig. A.1. Coordinate system of helical gears with nonparallel and nonintersecting AxesTh e procedure starts with a given, or generating surface r1(t, θ) for which a meshing, or generated surface is to be determined. A family of such gener-ated surfaces is given in parametric form by: r2(t, θ, τ ), where t is a profile parameter while θ and τ ar e motion parameters.r 1 =r 1(t, θ)=[ x 1,y 1,z 1]=x 01cosθ-y 01 sinθ, x 01 sinθ+ y 01 cosθ,p 1θ] (A,.1) ⎥⎦⎤⎢⎣⎡∂∂∂∂=∂∂0,,111t y t x t r=⎥⎦⎤⎢⎣⎡∂∂+∂∂∂∂-∂∂0,cos sin ,sin cos 0101011θθθθt y t x t y t x (A.2) []0,,0,,01010111x y y x r -=⎥⎦⎤⎢⎣⎡∂∂∂∂=∂∂θθθ (A.3) [][]∑+∑∑-∑-===c o s s i n ,s i n c o s ,,,),,(1111122222z y z y C x z y x t r r τθ []202020202,sin sin ,sin cos p y x y x ττττ+-= (A.4) [][]2020202022222,sin cos ,sin sin ,,p y x y x p x y r τττττ-+=-=∂∂ []∑--∑∑-+∑∑-∑=s i n )(c o s ,c o s )(s i n ,c o s s i n 121211C x p C x p y p θ (A.5) The envelope equation, which determines meshing between the surfaces r1 and r2:0222=∂∂∙⎪⎭⎫ ⎝⎛∂∂⨯∂∂τθr r t r (A.6) together with equations for these surfaces, completes a system of equations. If a generating surface 1 is defined by the parameter t, the envelope may be used to calculate another parameter θ, now a function of t, as a meshing condition to define a generated surface 2, now the function of both t and θ. The cross product in the envelope equation represents a surface normal and ∂r2 ∂τ is the relative, sliding velocity of two single points on the surfaces 1 and 2 which together form the common tangential point of contact of these two surfaces. Since the equality to zero of a scalar triple product is an invariant property under the applied coordinate system and since the relative velocity may be concurrently represented in both coordinate systems, a convenient form of the meshing condition is defined as:0211111=∂∂∙⎪⎭⎫ ⎝⎛∂∂⨯∂∂-=∂∂∙⎪⎭⎫ ⎝⎛∂∂⨯∂∂τθθθr r t r r r t r (A.7) Insertion of previous expressions into the envelope condition gives:[]⎪⎭⎫ ⎝⎛∂∂+∂∂∑-+-t y y t x x p p x C 1111211cot )( 0)cot (12111=⎥⎦⎤⎢⎣⎡∂∂∑-+∂∂+t x C p t y p p θ (A.8) This is applied here to derive the condition of meshing action for crossed helical gears of uniform lead with nonparallel and nonintersecting axes. The method constitutes a gear generation procedure which is generally applicable. It can be used for synthesis purposes of screw compressor rotors, which are electively helical gears with parallel axes. Formed tools for rotor manufacturing are crossed helical gears on non parallel and non intersecting axes with a uniform lead, as in the case of hobbing, or with no lead as in formed milling and grinding. Templates for rotor inspection are the same as planar rotor hobs. In all these cases the tool axes do not intersectthe rotor axes.Accordingly the notes present the application of the envelope method to produce a meshing condition for crossed helical gears. The screw rotor gearing is then given as an elementary example of its use while a procedure for forming a hobbing tool is given as a complex case.The shaft angle Σ, centre dist ance C, and unit leads of two crossed helical gears, p1 and p2 are not interdependent. The meshing of crossed helical gears is still preserved: both gear racks have the same normal cross section profile, and the rack helix angles are related to the shaft angle as Σ = ψr1+ ψr2. This is achieved by the implicit shift of the gear racks in the x direction forcing them to adjust accordingly to the appropriate rack helix angles. This certainly includes special cases, like that of gears which may be orientated so that the shaft angle is equal to the sum of the gear helix angles: Σ = ψ1+ ψ2. Furthermore a centre distance may be equal to the sum of the gear pitch radii :C = r1+ r2.Pairs of crossed helical gears may be with either both helix angles of the same sign or each of opposite sign, left or right handed, depending on the combination of their lead and shaft angle Σ. The meshing condition can be solved only by numerical methods. For the given parameter t, the coordinates x01 and y01 and their derivatives ∂x01∂t and ∂y01∂t are known. A guessed value of parameter θ is then used to calculate x1, y1, ∂x1 ∂t and ∂y1∂t. A revised value of θ is then derived and the procedure repeated until the difference between two consecutive values becomes sufficiently small.For give n transverse coordinates and derivatives of gear 1 profile, θ can be used to calculate the x1, y1, and z1 coordinates of its helicoid surfaces. The gear 2 helicoid surfaces may then be calculated. Coordinate z2 can then be used to calculate τ and finally, its transverse profile point coordinates x2, y2 can be obtained.A number of cases can be identified from this analysis.(i) When Σ = 0, the equation meets the meshing condition of screw machine rotors and also helical gears with parallel axes. For such a case, the gear helix angles have the same value, but opposite sign and the gear ratio i = p2/p1 is negative. The same equation may also be applied for the gen-eration of a rack formed from gears. Additionally it describes the formed planar hob, front milling tool and the template control instrument.122 A Envelope Method of Gearing(ii) If a disc formed milling or grinding tool is considered, it is suffcient to place p2= 0. This is a singular case when tool free rotation does not affect the meshing process. Therefore, a reverse transformation cannot be obtained directly.(iii) The full scope of the meshing condition is required for the generation of the profile of a formed hobbing tool. This is therefore the most compli-cated type of gear which can be generated from it.BReynolds Transport TheoremFollowing Hanjalic, 1983, Reynolds Transport Theorem defines a change of variable φ in a control volume V limited by area A of which vector the local normal is dA and which travels at local speed v. This control volume may, but need not necessarily coincide with an engineering or physical material system. The rate of change of variable φ in time within the volume is:⎰∂∂=⎪⎭⎫ ⎝⎛∂∂vV dV t t ρφφ (B.1) Therefore, it may be concluded that the change of variable φ in the volume V is caused by: – change of the specific variable m /φϕ=in time within the volume because of sources (and sinks) in the volume, ⎪⎭⎫ ⎝⎛∂∂t ϕdV which is called a local change and – movement of the control volume which takes a new space with variable ϕ in it and leaves its old space, causing a change in time of ϕfor ρϕv.dA and which is called convective changeThe first contribution may be represented by a volume integral:.()dV t V⎰∂∂ρϕ (B.2) while the second contribution may be represented by a surface integral:⎰⋅AdA V ρϕ (B.3) Therefore:()⎰⎰⎰⋅+∂∂==⎪⎭⎫ ⎝⎛∂∂AV V V dA V dV t dV dt d t ρϕρϕρϕφ ( B.4) which is a mathematical representation of Reynolds Transport Theorem.Applied to a material system contained within the control volume V m which has surface A m and velocity v which is identical to the fluid velocity w, Reynolds Transport Theorem reads:()dA W dV t d dt d t AmVm Vm Vm ⋅+∂∂==⎪⎭⎫ ⎝⎛∂∂⎰⎰⎰ρϕρϕρϕφV (B.5) If that control volume is chosen at one instant to coincide with the control volume V , the volume integrals are identical for V and Vm and the surface integrals are identical for A and Am , however, the time derivatives of these integrals are different, because the control volumes will not coincide in the next time interval. However, there is a term which is identical for the both timesintervals:()()dV t dV t VmV ⎰⎰∂∂=∂∂ρϕρϕ (B.6) therefore,⎰⎰⋅-⎪⎭⎫ ⎝⎛∂∂=⋅-⎪⎭⎫ ⎝⎛∂∂AV Am Vm dA v t dA w t ρϕφρϕφ (B.7) or:()dA v w t t AV Vm ⋅-+⎪⎭⎫ ⎝⎛∂∂=⎪⎭⎫ ⎝⎛∂∂⎰ρϕφφ (B.8) If the control volume is fixed in the coordinat e system, i.e. if it does not move, v = 0 and consequently:()dV tt V V ⎰∂∂=⎪⎭⎫ ⎝⎛∂∂ρϕφ (B.9) therefore:()⎰⎰⋅+∂∂=⎪⎭⎫ ⎝⎛∂∂AV Vm dA w dV t t ρϕρϕφ (B.10) Finally application of Gauss theorem leads to the common form:()()dV w dV t t VV Vm ⎰⎰⋅∇+∂∂=⎪⎭⎫ ⎝⎛∂∂ρϕρϕφ (B.11) As stated before, a change of variable φ is caused by the sources q within the volume V and influences outside the volume. These effects may be proportional to the system mass or volume or they may act at the system surface.The first effect is given by a volume integral and the second effect is given by a surface integral. ()⎰⎰⎰⎰=⋅∇+=⋅+=⎪⎭⎫ ⎝⎛∂∂VV A Vm Am A Vm qdV dV q qv dA q qvdV t φ (B.12) q can be scalar, vector or tensor.The combination of the two last equations gives:()⎰⎰⎰=⋅+∂∂A VV qdV dA w dV t ρϕρϕ Or:()()0=⎥⎦⎤⎢⎣⎡-⋅∇+∂∂⎰dV q w t V ρϕρϕ (B.13) Omitting integral signs gives:()()0=-⋅∇+∂∂q w tρϕρϕ (B.14)This is the well known conservation law form of variable ρϕφ=. Since for ϕ = 1, this becomes the continuity equation: ()0=⋅∇+∂∂w tρρ finally it is: ()()0=-∇⋅+⎥⎦⎤⎢⎣⎡⋅∇+∂∂q w w t ϕρρρϕ Or: ()q w tdt D =∇⋅+∂∂=ϕρϕρϕ (B.15) dt D /ϕ is the material or substantial derivative of variable ϕ. This equation is very convenient for the derivation of particular conservation laws. As previously mentioned ϕ = 1 leads to the continuity equation, ϕ = u to the momentum equation, ϕ = e, where e is specific internal energy, leads to the energy equation, ϕ = s, to the entropy equation and so on.If the surfaces, where the fluid carrying var iable Φ enters or leaves the control volume, can be identified, a convective change may conveniently be written:∙∙∙-∙-∙Φ-Φ=-==⋅⎰⎰out in out in A m m m d dA w )()(ϕϕϕϕρ (B.16) where the over scores indicate the variable average at entry/exit surface sections. This leads to the macroscopic form of the conservation law:()Q m m Q dt d dt d out in out in VV +-=+Φ-Φ=⎥⎦⎤⎢⎣⎡=⎪⎭⎫ ⎝⎛Φ∙-∙-∙∙)()(ϕϕρϕ (B.17) which states in words: (rate of change of Φ) = (inflow Φ) − (outflow Φ) +(source of Φ)中文译文A包络法的资产负债螺杆压缩机转子Stosic 1998年之后,被视为非平行不相交的螺旋齿轮,或在图的交叉轴。
中英文文献翻译-螺杆式压缩机
英文原文Screw CompressorsThe direction normal to the helicoids, can be used to calculate the coordinates of the rotorhelicoids n x and n y from x and y to which the clearance is added as:dt dyD p x x n δ+=, dt dxD p y y n δ-=, ⎪⎭⎫ ⎝⎛+=dt dy y dt dx x D z n δ (2.19) where the denominator D is given as :22222⎪⎭⎫ ⎝⎛++⎪⎭⎫ ⎝⎛+⎪⎭⎫ ⎝⎛=dt dy y dt dx x dt dy p dt dx p x D (2.20) n x and n y serve to calculate new rotor end plane coordinates, x 0n and y 0n ,with the clearances obtained for angles θ = n z /p and τ respectively. These on x and on y now serve to calculate the transverse clearance δ0 as the difference between them, as well as the original rotor coordinates o x and o y .If by any means, the rotors change their relative position, the clearance distribution at one end of the rotors may be reduced to zero on the flat side of the rotor lobes. In such a case, rotor contact will be prohibitively long on the flat side of the profile, where the dominant relative rotor motion is sliding, as shown in Fig. 2.29. This indicates that rotor seizure will almost certainly occur in that region if the rotors come into contact with each other.Fig. 2.29. Clearance distribution between the rotors: at suction, mid rotors, and discharge withpossible rotor contact at the dischargeFig. 2.30. Variable clearance distribution applied to the rotors It follows that the clearance distribution should be non-uniform to avoid hard rotor contact in rotor areas where sliding motion between the rotors is dominant.In Fig. 2.30, a reduced clearance of 65 μm is presented, which is now applied in rotor regions close to the rotor pitch circles, while in other regions it is kept at 85 μm, as was done by Edstroem, 1992. As can be seen in Fig. 2.31, the situation regarding rotor contact is now quite different. This is maintained along the rotor contact belt close to the rotor pitch circles and fully avoided at other locations. It follows that if contact occurred, it would be of a rolling character rather than a combination of rolling and sliding or even pure sliding. Such contact will not generate excessive heat and could therefore be maintained for a longer period without damaging the rotors until contact ceases or the compressor is stopped.2.6 Tools for Rotor ManufactureThis section describes the generation of formed tools for screw compressor hobbing, milling and grinding based on the envelope gearing procedure.2.6.1 Hobbing ToolsA screw compressor rotor and its formed hobbing tool are equivalent to a pair of meshing crossed helical gears with nonparallel and nonintersecting axes. Their general meshing condition is given in Appendix A. Apart from the gashes forming the cutter faces, the hob is simply a helical gear in which.Fig. 2.31. Clearance distribution between the rotors: at suction, mid of rotor and discharge with apossible rotor contact at the dischargeEach referred to as a thread, Colburne, 1987. Owing to their axes not being parallel, there is only point contact between them whereas there is line contact between the screw machine rotors. The need to satisfy the meshing equation given in Appendix A, leads to the rotor – hob meshing requirement for the given rotor transverse coordinate points 1o x and 1o y and their first derivative 0101dx dy .The hob transverse coordinate points 2o x and 2o y can then be calculated. These are sufficient to obtain the coordinate 2012012y x R +=The axial coordinate 2z , calculated directly, and 2R are hob axial plane coordinates which define the hob geometry.The transverse coordinates of the screw machine rotors, described in the previous section, are used as an example here to produce hob coordinates. he rotor unit leads 1P are 48.754mm for the main and −58.504mm for the ate rotor. Single lobe hobs are generated for unit leads 2P :6.291mm for the m ain rotor and −6.291mm for the gate rotor. The corresponding hob helix a ngles ψ are 85◦ and 95◦. The same rotor-to-hob centre distance C = 110mm a nd the shaft angle Σ = 50◦ are given for both rotors. Figure 2.32 contains a view to the hob.Reverse calculation of the hob – screw rotor transformation, also given in Appendix Apermits the determination of the transverse rotor profile coordinates which will be obtained as a result of the manufacturing process. These ay be compared with those originally specified to determine the effect ofFig. 2.32. Rotor manufacturing: hobbing tool left , right milling toolmanufacturing errors such as imperfect tool setting or tool and rotor deformation upon the final rotor profile.For the purpose of reverse transformation, the hob longitudinal plane coordinates 2R and 2z and 22dz dR should be given. The axial coordinate 2z is used to calculate 22P Z T =, which is then used to calculate the hob transverse coordinates:τcos 202R x =, τs i n 202R y = (2.21)These are then used as the given coordinates to produce a meshing criterionand the transverse plane coordinates of the “manufactured” rotors.A comparison between the original rotors and the manufactured rotors is given in Fig. 2.33 with the difference between them scaled 100 times. Two types of error are considered. The left gate rotor, is produced with 30um offset in the centre distance between the rotor and the tool, and the main rotor withFig. 2.33. Manufacturing imperfections0.2◦ of fset in the tool shaft angle Σ. Details of this particular meshing method are given by Stosic 1998.2.6.2 Milling and Grinding ToolsFormed milling and grinding tools may also be generated by placing 02=P in the general meshing equation, given in Appendix A, and then following the procedure of this section. The resulting meshing condition now reads as:[]0cot cot 1111111111=⎥⎦⎤⎢⎣⎡∂∂-∂∂+⎪⎭⎫ ⎝⎛∂∂+∂∂∑+-∑t x C t y p p t y y t x x p x C θ (2.22) However in this case, when one expects to obtain screw rotor coordinates from the tool coordinates, the singularity imposed does not permit the calculation of the tool transverse plane coordinates. The main meshing condition cannot therefore be applied. For this purpose another condition is derived for the reverse milling tool to rotor transformation from which the meshing angle τ is calculated:()0cot sin cot cos 12212222=-∑+∑++⎪⎪⎭⎫ ⎝⎛+C p dR dz C p dR dz z R ττ (2.23) Once obtained, τ will serve to calculate the rotor coordinates after the “manufacturing” process. The obtained rotor coordinates will contain all manufacturing imperfections, like mismatch of the rotor – tool centre distance, error in the rotor – tool shaft angle, axial shift of the tool or tool deformation during the process as they are input to the calculation process. A full account of this useful procedure is given by Stosic 1998.2.6.3 Quantification of Manufacturing ImperfectionsThe rotor – tool transformation is used here for milling tool profile generation. The reverse procedure is used to calculate the “manufactured” rotors. The rack generated 5-6 128mm rotors described by Stosic, 1997a are used as given profiles: x (t ) and y (t ). Then a tool – rotor transformation is used to quantify the influence of manufacturing imperfections upon the qualityof the produced rotor profile. Both, linear and angular offset were considered.Figure 2.33 presents the rotors, the main manufactured with the shaft angle offset 0.5◦and the gate with the centre distance offset 40 μm from that of the original rotors given by the dashed line on the left. On the right, the rotors are manufactured with imperfections, the main with a tool axial offset of 40 μm and the gate with a certain tool body deformation which resulted in 0.5◦offset of the relative motion angle θ. The original rotors are given by the dashed line.3Calculation of Screw Compressor Performance Screw compressor performance is governed by the interactive effects of thermodynamic and fluid flow processes and the machine geometry and thus can be calculated reliably only by their simultaneous consideration. This may be chieved by mathematical modelling in one or more dimensions. For most applications, a one dimensional model is sufficient and this is described in full. 3-D modelling is more complex and is presented here only in outline. A more detailed presentation of this will be made in a separate publication.3.1 One Dimensional Mathematical ModelThe algorithm used to describe the thermodynamic and fluid flow processes in a screw compressor is based on a mathematical model. This defines the instantaneous volume of the working chamber and its change with rotational angle or time, to which the conservation equations of energy and mass continuity are applied, together with a set of algebraic relationships used to define various phenomena related to the suction, compression and discharge of the working fluid. These form a set of simultaneous non-linear differential equations which cannot be solved in closed form.The solution of the equation set is performed numerically by means of the Runge-Kutta 4th order method, with appropriate initial and boundary conditions.The model accounts for a number of “real-life” effects, which may significantly influence the performance of a real compressor. These make it suitable for a wide range of applications and include the following:– The working fluid compressed can be any gas or liquid-gas mixture for which an equation of state and internal energy-enthalpy relation is known, i.e. any ideal or real gas or liquid-gas mixture of known properties.–The model accounts for heat transfer between the gas and the compressor rotors or its casing in a form, which though approximate, reproduces the overall effect to a good first order level of accuracy.– The model accounts for leakage of the working medium through the clearances between the two rotors and between the rotors and the stationary parts of the compressor.– The process equations and the subroutines for their solution are independent of those which define the compressor geometry. Hence, the model can be readily adapted to estimate the performance of any geometry or type of positive displacement machine.– The effects of liquid injection, including that of oil, water, or refrigerant can be accounted for during the suction, compression and discharge stages.– A set of subroutines to estimate the thermodynamic properties and changes of state of the working fluid during the entire compressor cycle of operations completes the equation set and thereby enables it to be solved.Certain assumptions had to be introduced to ensure efficient computation.These do notimpose any limitations on the model nor cause significant departures from the real processes and are as follows:– The fluid flow in the model is assumed to be quasi one-dimensional.–Kinetic energy changes of the working fluid within the working chamber are negligible compared to internal energy changes.–Gas or gas-liquid inflow to and outflow from the compressor ports is assumed to be isentropic.– Leakage flow of the fluid through the clearances is assumed to be adiabatic.3.1.1 Conservation EquationsFor Control Volume and Auxiliary RelationshipsThe working chamber of a screw machine is the space within it that contains the working fluid. This is a typical example of an open thermodynamic system in which the mass flow varies with time. This, as well as the suction and discharge plenums, can be defined by a control volume for which the differential equations of the conservation laws for energy and mass are written. These are derived in Appendix B, using Reynolds Transport Theorem.A feature of the model is the use of the non-steady flow energy equation to compute the thermodynamic and flow processes in a screw machine in terms of rotational angle or time and how these are affected by rotor profile modifications. Internal energy, rather than enthalpy, is then the derived variable. This is computationally more convenient than using enthalpy as the derived Variable since, even in the case of real fluids, it may be derived, without reference to pressure. Computation is then carried out through a series of iterative cycles until the solution converges. Pressure, which is the desired output variable, can then be derived directly from it, together with the remaining required thermodynamic properties.The following forms of the conservation equations have been employed in the model:中文翻译螺杆式压缩机几何的法线方向的螺旋,可以用来计算的坐标转子螺旋n x 和n y 的从x 和y 的间隙加入如:dt dyD p x x n δ+=, dt dxD p y y n δ-=, ⎪⎭⎫ ⎝⎛+=dt dy y dt dx x D z n δ (2.19) 其中分母D 被给定为:22222⎪⎭⎫ ⎝⎛++⎪⎭⎫ ⎝⎛+⎪⎭⎫ ⎝⎛=dt dy y dt dx x dt dy p dt dx p x D (2.20) n x ,n y 服务来计算新的转子端的平面的坐标,on x 和on y ,得到的间隙角θ =锌/ p 和τ 。
螺杆式压缩机的设计外文文献翻译、中英文翻译、外文翻译
英文原文1 IntroductionThe screw compressor is one of the most common types of machine used to compress gases. Its construction is simple in that it essentially comprises only a pair of meshing rotors, with helical grooves machined in them, contained in a casing, which fits closely round them. The rotors and casing are separated by very small clearances. The rotors are driven by an external motor and mesh like gears in such a manner that, as they rotate, the space formed between them and the casing is reduced progressively. Thus, any gas trapped in this case is compressed. The geometry of such machines is complex and the flow of the gas being compressed within them occurs in three stages. Firstly, gas enters between the lobes, through an inlet port at one end of the casing during the start of rotation. As rotation continues, the space between the rotors no longer lines up with the inlet port and the gas is trapped and thus compressed. Finally, after further rotation, the opposite ends of the rotors pass a second port at the other end of the casing, through which the gas is discharged. The whole process is repeated between successive pairs of lobes to create a continuous but pulsating flow of gas from low to high pressure.These machines are mainly used for the supply of compressed air in the building industry, the food, process and pharmaceutical industries and, where required, in the metallurgical industry and for pneumatic transport.They are also used extensively for compression of refrigerants in refrigeration and air conditioning systems and of hydrocarbon gases in the chemical industry. Their relatively rapid acceptance over the past thirty years is due to their relatively high rotational speeds compared to other types of positive displacement machine, which makes them compact, their ability to maintain high efficiencies over a wide range of operating pressures and flow rates and their long service life and high reliability. Consequently, they constitute a substantial percentage of all positive displacement compressors now sold and currently in operation.The main reasons for this success are the development of novel rotor profiles, which have drastically reduced internal leakage, and advanced machine tools, which can manufacture the most complex shapes to tolerances of the order of 3 micrometers at an acceptable cost. Rotor profile enhancement is still the most promising means of further improving screw compressors and rational procedures are now being developed both to replace earlier empirically derived shapes and also to vary the proportions of the selected profile to obtain the best result for the application for which the compressor is required. Despite their wide usage, due to the complexity of their internal geometry and the non-steady nature of the processes within them, up till recently, only approximate analytical methods have been available to predict their performance. Thus, although it is known that their elements are distorted both by the heavy loads imposed by pressure induced forces and through temperature changes within them, no methods were available to predict the magnitude of these distortions accurately, nor how they affect the overall performance of the machine. In addition, improved modelling of flow patterns within the machine can lead to better porting design. Also, more accurate determination of bearing loads and how they fluctuate enable better choices of bearings to be made. Finally, if rotor and casing distortion, as a result of temperature and pressure changes within the compressor, can be estimated reliably, machining procedures can be devised to minimise their adverse effects.Screw machines operate on a variety of working fluids, which may be gases, dry vapour or multi-phase mixtures with phase changes taking place within the machine. They may involve oil flooding, or other fluids injected during the compression or expansion process, or be without any form of internal lubrication. Their geometry may vary depending on the number of lobes in each rotor, the basic rotor profile and the relative proportions of each rotor lobe segment. It follows that there is no universal configuration which would be the best for all applications. Hence, detailed thermodynamic analysis of the compression process and evaluation of the influence of the various design parameters on performance is more important to obtain the best results from these machines than from other types which could be used for the same application. A set of well defined criteria governed by an optimisation procedure is therefore a prerequisite for achieving the best design for each application. Such guidelines are also essential for the further improvement of existing screw machine designs and broadening their range of uses. Fleming et al., 1998 gives a good contemporary review of screw compressor modelling, design and application.A mathematical model of the thermodynamic and fluid flow processes within positive displacement machines, which is valid for both the screw compressor and expander modes of operation, is presented in this Monograph. It includes the use of the equations of conservation of mass, momentum and energy applied to an instantaneous control volume of trapped fluid within the machine with allowance for fluid leakage, oil or other fluid injection, heat transfer and the assumption of real fluid properties. By simultaneous solution of these equations, pressure-volume diagrams may be derived of the entire admission, discharge and compression or expansion process within the machine. A screw machine is defined by the rotor profile which is here generated by use of a general gearing algorithm and the port shape and size. This algorithm demonstrates the meshing condition which, when solved explicitly,enables a variety of rotor primary arcs to be defined either analytically or by discrete point curves. Its use greatly simplifies the design since only primary arcs need to be specified and these can be located on either the main or gate rotor or even on any other rotor including a rack, which is a rotor of infinite radius. The most efficient profiles have been obtained from a combined rotor-rack generation procedure.The rotor profile generation processor, thermofluid solver and optimizer,together with pre-processing facilities for the input data and graphical post processing and CAD interface, have been incorporated into a design tool in the form of a general computer code which provides a suitable tool for analysis and optimization of the lobe profiles and other geometrical and physical parameters. The Monograph outlines the adopted rationale and method of modelling, compares the shapes of the new and conventional profiles and illustrates potential improvements achieved with the new design when applied to dry and oil-flooded air compressors as well as to refrigeration screw compressors.The first part of the Monograph gives a review of recent developments in screw compressors.The second part presents the method of mathematical definition of the general case of screw machine rotors and describes the details of lobe shape specification. It focuses on a new lobe profile of a slender shape with thinner lobes in the main rotor, which yields a larger cross-sectional area and shorter sealing lines resulting in higher delivery rates for the same tip speed.The third part describes a model of the thermodynamics of the compression-expansion processes, discusses some modelling issues and compares the shapes of new and conventional profiles. It illustrates the potentialimprovements achievable with the new design applied to dry and oil-flooded air compressors as well as to refrigeration screw compressors. The selection of the best gate rotor tip radius is given as an example of how mathematical modelling may be used to optimise the design and the machine’s operating conditions.The fourth part describes the design of a high efficiency screw compressor with new rotor profiles. A well proven mathematical model of the compression process within positive displacement machines was used to determine the optimum rotor size and speed, the volume ratio and the oil injection position and jet diameter. In addition, modern design concepts such as an open suction port and early exposure of the discharge port were included, together with improved bearing and seal specification, to maximise the compressor efficiency. The prototypes were tested and compared with the best compressors currently on the market. The measured specific power input appeared to be lower than any published values for other equivalent compressors currently manufactured. Both the predicted advantages of the new rotor profile and the superiority of the design procedure were thereby confirmed.1.1 Basic ConceptsThermodynamic machines for the compression and expansion of gases and vapours are the key components of the vast majority of power generation and refrigeration systems and essential for the production of compressed air and gases needed by industry. Such machines can be broadly classified by their mode of operation as either turbomachines or those of the positive displacement type.Turbomachines effect pressure changes mainly by dynamic effects, related to the change of momentum imparted to the fluids passing through them. These are associated with the steady flow of fluids at high velocities and hence these machines are compact and best suited for relatively large mass flow rates. Thus compressors and turbines of this type are mainly used in the power generation industry, where, as a result of huge investment in research and development programmes, they are designed and built to attain thermodynamic efficiencies of more than 90% in large scale power production plant. However, the production rate of machines of this type is relatively small and worldwide, is only of the order of some tens of thousands of units per annum.Positive displacement machines effect pressure changes by admitting a fixed mass of fluid into a working chamber where it is confined and then compressed or expanded and, from which it is finally discharged. Such machines must operate more or less intermittently. Such intermittent operation is relatively slow and hence these machines are comparatively large. They are therefore better suited for smaller mass flow rates and power inputs and outputs. A number of types of machine operate on this principle such as reciprocating, vane, scroll and rotary piston machines.In general, positive displacement machines have a wide range of application, particularly in the fields of refrigeration and compressed air production and their total world production rate is in excess of 200 million units per annum. Paradoxically, but possibly because these machines are produced by comparatively small companies with limited resources, relatively little is spent on research and development programmes on them and there are very few academic institutions in the world which are actively promoting their improvement.One of the most successful positive displacement machines currently in use is the screw or twin screw compressor. Its principle of operation, as indicated in Fig. 1.1, is based on volumetric changes in three dimensions rather than two. As shown, it consists, essentially, of a pair of meshing helical lobed rotors, contained in a casing.The spaces formed between the lobes on each rotor form a series of working chambers in which gas or vapour is contained. Beginning at the top and in front of the rotors, shown in the light shaded portion of Fig. 1.1a, there is a starting point for each chamber where the trapped volume is initially zero. As rotation proceeds in the direction of the arrows, the volume of that chamber then increases as the line of contact between the rotor with convex lobes, known as the main rotor, and the adjacent lobe of the gate rotorFig. 1.1. Screw Compressor Rotorsadvances along the axis of the rotors towards the rear. On completion of one revolution i.e. 360◦by the main rotor, the volume of the chamber is then a maximum and extends in helical form along virtually the entire length of the rotor. Further rotation then leads to reengagement of the main lobe with the succeeding gate lobe by a line of contact starting at the bottom and front of the rotors and advancing to the rear, as shown in the dark shaded portions in Fig. 1.1b. Thus, the trapped volume starts to decrease. On completion of a further 360◦of rotation by the main rotor, the trapped volume returns to zero.The dark shaded portions in Fig. 1.1 show the enclosed region where therotors are surrounded by the casing, which fits closely round them, while the light shaded areas show the regions of the rotors, which are exposed to external pressure. Thus the large light shaded area in Fig. 1.1a corresponds to the low pressure port while the small light shaded region between shaft ends B and D in Fig. 1.1b corresponds to the high pressure port.Exposure of the space between the rotor lobes to the suction port, as their front ends pass across it, allows the gas to fill the passages formed between them and the casing until the trapped volume is a maximum. Further rotation then leads to cut off of the chamber from the port and progressive reduction in the trapped volume. This leads to axial and bending forces on the rotors and also to contact forces between the rotor lobes. The compression process continues until the required pressure is reached when the rear ends of the passages are exposed to the discharge port through which the gas flows out at approximately constant pressure. It can be appreciated from examination of Fig. 1.1, is that if the direction of rotation of the rotors is reversed, then gas will flow into the machine through the high pressure port and out through the low pressure port and it will act as an expander. The machine will also work as an expander when rotating in the same direction as a compressor provided that the suction and discharge ports are positioned on the opposite sides of the casing to those shown since this iseffectively the same as reversing the direction of rotation relative to the ports. When operating as a compressor, mechanical power must be supplied to shaft A to rotate the machine. When acting as an expander, it will rotate automatically and power generated within it will be supplied externally through shaft A.The meshing action of the lobes, as they rotate, is the same as that of helical gears but, in addition, their shape must be such that at any contact position, a sealing line is formed between the rotors and between the rotors and the casing in order to prevent internal leakage between successive trapped passages. A further requirement is that the passages between the lobes should be as large as possible, in order to maximise the fluid displacement per revolution. Also, the contact forces between the rotors should be low in order to minimise internal friction losses.A typical screw rotor profile is shown in Fig. 1.2, where a configuration of 5–6 lobes on the main and gate rotors is presented. The meshing rotors are shown with their sealing lines, for the axial plane on the left and for the cross-sectional plane in the centre. Also, the clearance distribution between the two rotor racks in the transverse plane, scaled 50 times (6) is given above.Fig. 1.2. Screw rotor profile: (1) main, (2) gate, (3) rotor external and (4) pitch circles, (5) sealing line, (6) clearance distribution and (7) rotor flow area between the rotors and housingOil injected Oil FreeFig. 1.3. Oil Injected and Oil Free CompressorsScrew machines have a number of advantages over other positive displacement types. Firstly, unlike reciprocating machines, the moving parts all rotate and hence can run at much higher speeds. Secondly, unlike vane machines, the contact forces within them are low, which makes them very reliable. Thirdly, and far less well appreciated, unlike the reciprocating, scroll and vane machines, all the sealing lines of contact which define the boundaries of each cell chamber, decrease in length as the size of the working chamber decreases and the pressure within it rises. This minimises the escape of gas from the chamber due to leakage during the compression or expansion process.1.2 Types of Screw CompressorsScrew compressors may be broadly classified into two types. These are shown in Fig. 1.3 where machines with the same size rotors are compared:1.2.1 The Oil Injected MachineThis relies on relatively large masses of oil injected with the compressed gas in order to lubricate the rotor motion, seal the gaps and reduce the temperature rise during compression. It requires no internal seals, is simple in mechanical design, cheap to manufacture and highly efficient. Consequently it is widely used as a compressor in both the compressed air and refrigeration industries.1.2.2 The Oil Free MachineHere, there is no mixing of the working fluid with oil and contact between the rotors is prevented by timing gears which mesh outside the working chamber and are lubricated externally. In addition, to prevent lubricant entering the working chamber, internal seals are required on each shaft between the working chamber and the bearings. In the case of process gas compressors, double mechanical seals are used. Even with elaborate and costly systems such as these, successful internal sealing is still regarded as a problem by established process gas compressor manufacturers. It follows that such machines are considerably more expensive to manufacture than those that are oil injected.Both types require an external heat exchanger to cool the lubricating oil before it is readmitted to the compressor. The oil free machine requires an oil tank, filters and a pump to return the oil to the bearings and timing gear.The oil injected machine requires a separator to remove the oil from the high pressure discharged gas but relies on the pressure difference between suction and discharge to return the separated oil to the compressor. Theseadditional components increase the total cost of both types of machine but the add on cost is greater for the oil free compressor.1.3 Screw Machine DesignSerious efforts to develop screw machines began in the nineteen thirties, when turbomachines were relatively inefficient. At that time, Alf Lysholm, a talented Swedish engineer, required a high speed compressor, which could be coupled directly to a turbine to form a compact prime mover, in which the motion of all moving parts was purely rotational. The screw compressor appeared to him to be the most promising device for this purpose and all modern developments in these machines stem from his pioneering work. Typical screw compressor designs are presented in Figs. 1.4 and 1.5. From then until the mid nineteen sixties, the main drawback to their widespread use was the inability to manufacture rotors accurately at an acceptable cost. Two developments then accelerated their adoption. The first was the development of milling machines for thread cutting. Their use for rotor manufacture enabled these components to be made far more accurately at an acceptable cost. The second occurred in nineteen seventy three, when SRM, in Sweden, introduced the “A” profile, which reduced the internal leakage path area, known as the blow hole, by 90%. Screw compressors could then be built with efficiencies approximately equal to those of reciprocating machines and, in their oil flooded form, could operate efficiently with stage pressure ratios of up to 8:1. This was unattainable with reciprocating machines. The use of screw compressors, especially of the oil flooded type, then proliferated.Fig. 1.4. Screw compressor mechanical partsFig. 1.5. Cross section of a screw compressor with gear boxTo perform effectively, screw compressor rotors must meet the meshing requirements of gears while maintaining a seal along their length to minimise leakage at any position on the band of rotor contact. It follows that the compressor efficiency depends on both the rotor profile and the clearances between the rotors and between the rotors and the compressor housing.Screw compressor rotors are usually manufactured on pecialized machines by the use of formed milling or grinding tools. Machining accuracy achievable today is high and tolerances in rotor manufacture are of the order of 5 μm around the rotor lobes. Holmes, 1999 reported that even higher accuracy was achieved on the new Holroyd vitrifying thread-grinding machine, thus keeping the manufacturing tolerances within 3 μm even in large batch production. This means that, as far as rotor production alone is concerned, clearances betweenthe rotors canbe as small as 12 μm.中文译文1 引言螺杆式压缩机是一种最常见的用来压缩气体的机器。
单螺杆机床论文中英文对照资料外文翻译文献
中英文对照资料外文翻译文献Dedicated to the single screw compressor machine updated the IntroductionAbstract: This paper describes four areas from the existingsingle-screw machine layout and structure, and put out the advantages and disadvantages of the list, because of the compressor plant single-screw machine tools and machine tool external Security information, the above introduction there is inevitably one-sided and wrong, and are therefore single-screw compressor for the production of reference works.First, introduce the layout of machine toolsDecide the size of the compressor displacement of the stars round, screw diameter, mesh size and the size of the center distance, so different in diameter screw, machine tool spindle and the rotary center are also different. To meet the processing of different diameter screw, single screw Currently the layout of machine tools in general there are several options.The first is: machine tool rotary tool spindle center and the center distance for the fixedMachine tool rotary tool spindle center and the center distance for the fixed, can not adjust the center distance. Processing of several of the screw diameter on the center distance required several different specifications of the machine.Advantages: simple structure of the machine.Disadvantage: each machine can only process a specification of the screw, when the market on a certain specification requirements when the screw compressor, resulting in a machine, other machine idle.The second: the machine tool spindle box for rotaryProcessing screw machine according to the size of the diameter at the processing before a point of rotating spindle box. Spindle box that the machine can turn on a machine at the above-mentioned article on the use of the improvements, with the first structure of a machine tool is basically the same.Advantages: the structure of machine tool easy to adapt to a variety of specifications of the processing screw.One disadvantage: after the rotating spindle box and the tool spindle turning center line distance between the center line of accurate measurement difficult.2 disadvantage: after the rotating spindle spindle box and the front surface of the rotary cutter centerline distance between the reduction of the larger diameter of the screw processing is limited. The third: the machine tool spindle box for horizontal mobileBox at the bottom of the spindle and the base there is arranged between the rectangular sliding rail, spindle box perpendicular to the direction of movement of spindle centerline and perpendicular to the centerline of the tool rotation. Through the power of the spindle box spline shaft to the base of the tool feed mechanism. Screw diameter, according to the size of the processing in the processing of the previous round by hand to the body put into the screw spindle box moved to the appropriate location, and then screw the spindle box on a fixed base. Spindle box available from the mobile Grating detection, position error ± 0.005mm. Horizontal spindle box can be used as a mobile machine can process diameter φ95 ~ φ385mm any kind between the screw specifications.Φ95 ~ φ385mm processing because of the diameter of the screw, causing the front surface and the tool spindle rotation the distance between the center line of the margin is too large, the actual application in the design specifications of the machine into two, aφ95 ~ φ205mm machine screw diameter Another φ180 ~ φ385mm machine screw diameter.Advantages: a variety of tools to adapt to the specifications of the processing screw, each screw specifications need not be provided with the appropriate machine tools.Disadvantage: the structure of machine tools and machine tool assembly of the two kinds of more complex machine tools, machine tools than the cost of two kinds of machine tools before the high. Second, introduce the structure of machine tool spindleThe level of machine tool spindle box on the main axis and the base of the vertical axis determines the degree of precision was the precision screw machining, at the same time screw compressor at a speed of thousands of high-speed rotary switch, the accuracy of the screw will be less so that the compressor have a fever, vibration, low efficiency, such as wear and tear situation quickly.Currently available single-screw machine spindle structure of the program has the following two.The first is: bearing radial clearance is not adjustable spindle structureBefore spindle bearing out the use of one pairs of cylindrical roller bearings and thrust ball bearing combination of both, the main useof double row cylindrical roller bearings under radial cutting force, the use of two ball bearings to bear axial thrust cutting force.After the general adoption of the spindle bearings out one pairs of cylindrical roller bearings or a ball bearing to the heart.Main advantages of this structure: the main axis of the processing and assembly of simple, low cost.One disadvantage: because the main axis of the radial bearing clearance can not be adjusted so poor precision spindle. Although the use of bearings and shaft diameter fit to eliminate the radial bearing clearance, but each bearing diameter and radial clearance is not a fixed value, so it is difficult to design and processing to the quasi-axial-radial and bearings with bore tolerances.2 disadvantage: it is very difficult to buy in the market of domestically produced or imported, C, D or P4, P5 class thrust ball bearings, machine tool manufacturing plant commonly used alternative to the use of ordinary class bearings, which also affected the accuracy of the enhance spindle.Bearing radial clearance adjustable spindle structure do not apply to the general accuracy of the general machine tools, does not apply to require a higher accuracy of the spindle of machine tools. The second: the radial bearing clearance adjustable spindle structureBefore the adoption of a spindle bearing P4 class of double row tapered hole cylindrical roller bearings and a P4-class double row ball bearing thrust to the combination of heart. The use of the spindle hole of the double row tapered cylindrical roller bearings under radial cutting force, the use of double row ball bearing thrust to the heart to bear part of the axial and radial cutting force cutting force.Spindle bearings generally used after a P5 class of double row tapered hole cylindrical roller bearings.Double row tapered hole cylindrical roller bearings with inner ring and shaft are tapered 1:12, bearing lock nut with a round led a bearing in the axial displacement of the inner ring bearings and expansion, to reduce or eliminate Bearing radial clearance purposes.Main structure of such advantages: high precision spindle. At the front spindle diameter φ230mm noodle on the end measuring spindle Beat value of 0.010mm. Φ230mm cylindric al spindle at the front end on the radial axis measurement value of Beat 0.005mm. The second structure of the spindle of a precision spindle accuracy than the first about 50% improve.Main disadvantage of this structure:The principal axis of the more complicated process, the spindle assembly also has the experience necessary to make the workers to operate the spindle achieve the desired numerical accuracy. Third, the depth of the tool feed controlRequired different processing screw diameter spiral groove depth is also different from the depth of the spiral groove mm from dozens to more than 100 millimeters range around the tool into the institutions required to feed the thousands of ring rotation in order to achieve a screw machining .Feed because of the tool in the tool rotating at the same time achieve motion feed, so on a number of general machine tools used in mechanical, electrical control method of depth of cut does not apply to single-screw machine.Single screw machine tools give agencies into the following different methods can be feed to control the depth of purpose.The first is: friction clutch and electrical switches to control the depth of the tool feedIts principle is to control depth of cut increases the tool cutter feed mechanism increases the load torque so that the tool feeding mechanism of the friction transmission chain slipping clutch, a mechanical linkage concurrent silent trigger electrical switches,optical signal prompted operator, when manual operator to disconnect the tool into the power sector.The advantages of this control method are: the control method is simple and spare parts processing and operational power from the impact of a sudden.Disadvantage are: processing of different diameter screw to adjust the clutch friction discs pressed the preload spring.Material because of the density of each screw, and the hardness of the existence of subtle differences in the degree of cutting tools sharp differences exist, thus the accuracy of this control method was not too accurate, may lead to screw spiral groove depth tolerance is too large.The second: use of an electromagnetic clutch, encoder control tool into the mix to the depth ofTool feed system, equipped with electromagnetic clutch and a tool for detecting the number of rotating ring gear and a gun encoder. It is a tool of control principle hand screw surface encoder to start counting switch, then start counting counting device, when the rotary tool to pre-set number of laps when the cutting depth is reached, the electromagnetic clutch automatic off open to the power tool into the concurrent silent, optical signal parts prompted the operator has finished processing.The detection device through the digital display shows the number of feed circles or feed. Torn off and the electromagnetic clutch, the tool does not only into the rotation with the vertical shaft to the sport.The advantages of this control method are: the depth of the spiral groove screw tolerance control more accurate, because of several significant table shows the depth of processing, or want a few laps and the depth of processing or circle the number of operations is also very intuitive and user-friendly.Disadvantage are: electrical control of machine tools at the same time more complex parts of this control method at the processing plant, if a sudden power failure, the prior data set will be lost.If you add in the electrical control of the battery to power at the early-dimensional detection devices to maintain the job, the problem can be resolved.Four, the control gear drive spaceSingle screw machine screw in the processing, due to the spiral groove in the rotary tool and the workpiece rotation to complete the synthesis process. Just cut into the workpiece when the tool in the tangential direction of rotation has been going on a greater resistance knife, cutting tool at the workpiece to be cut when therole of the spiral groove, the tool in the tangential direction of rotation has been going up against a smaller knife and even by the spiral groove thrust workpiece.Because there is a box-hole processing machine tool, gear and other processing error, the tool axis of rotation of the drive space is too large, large amount of so-called open.Detect drive way too much space is a fixed power input shaft and output shaft rotation shaking, in the case of the transmission structure of conventional design and manufacture of machine tools, the transmission output shaft angle space at more than ten degrees to the dozens of degrees. Transmission gap caused by too large spiral screw groove surface then there is obvious marks, thus affecting the machining accuracy of the screw.Upon completion of the assembly machine tool axis of rotation of the drive space is too large, in fact are subject to various errors gear, creating a backlash of the gear is too large.Machine tools in the mechanical transmission gear are used regardless of the accuracy of a few of the class, the designers take into account the gear manufacturing error, processing error box center distance, temperature, lubricating oil film thickness, the assembly error and other factors, machine design must ensure thattransmission gear A certain amount of backlash, backlash decide the size of the gear tooth thickness tolerance size.Single-screw machine has the Main Drive from other machine tool structure specificity. In order to reduce transmission or reasonable gap single-screw machine tools currently used by the following two ways.The first is: the installation at the output shaft brakeTool at the output shaft rotating the location of cylindrical symmetry with radial brake, brake stand up to the tool front-end of the cylindrical rotary output shaft, brake for spring preload.The working principle of the brake is generated by the friction brake to increase the output shaft damping, reducing the sensitivity of the rotation axis.Are: brake and easy does not change the structure of the original machine tool structure, the method of indirect reduction to achieve the purpose of drive space, in practical applications there is a certain effect.One disadvantage: the pre-spring brake tool because of the cylindrical output shaft to exert a greater radial force, in fact increases the load machine torque, resulting in increased motor power at the same time gears, bearings to accelerate wear and tear.Disadvantage 2: pre-spring brake because of the output shaft of the cylindrical tool to exert a greater radial force on the possible geometry of the tool output shaft a negative impact on accuracy.Conclusion: This article describes four areas from existingsingle-screw machine layout and structure, and put out the advantages and disadvantages of the list, because of the compressor plant single-screw machine tools and machine tool external Security information, the above introduction there is inevitably one-sided and wrong, and are therefore single-screw compressor for the production of reference works.对压缩机单螺杆专用加工机床的介绍更新时间摘要:本文从四个方面介绍了国内现有单螺杆加工机床的布局和结构,并把优缺点一一列举出来,由于压缩机生产厂的单螺杆加工机床和机床资料对外保密,以上介绍难免有片面、不妥之处,因此仅供单螺杆压缩机生产厂参考。
螺杆压缩机外文文献翻译、中英文翻译、外文翻译
螺杆压缩机外文文献翻译、中英文翻译、外文翻译英文原文Screw CompressorsN. Stosic I. Smith A. KovacevicScrew CompressorsMathematical Modellingand Performance CalculationWith 99 FiguresABCProf. Nikola StosicProf. Ian K. SmithDr. Ahmed KovacevicCity UniversitySchool of Engineering and Mathematical SciencesNorthampton SquareLondonEC1V 0HBU.K.e-mail:n.stosic@/doc/d6433edf534de518964bcf 84b9d528ea81c72f87.htmli.k.smith@/doc/d6433edf534de51896 4bcf84b9d528ea81c72f87.htmla.kovacevic@/doc/d6433edf534de51 8964bcf84b9d528ea81c72f87.htmlLibrary of Congress Control Number: 2004117305ISBN-10 3-540-24275-9 Springer Berlin Heidelberg New York ISBN-13 978-3-540-24275-8 Springer Berlin Heidelberg New YorkThis work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable for prosecution under the GermanCopyright Law.Springer is a part of Springer Science+Business Media/doc/d6433edf534de518964bcf84b9d 528ea81c72f87.html_c Springer-Verlag Berlin Heidelberg 2005Printed in The NetherlandsThe use of general descriptive names, registered names, trademarks, etc. in this publication does not imply,even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.Typesetting: by the authors and TechBooks using a Springer LATEX macro packageCover design: medio, BerlinPrinted on acid-free paper SPIN: 11306856 62/3141/jl 5 4 3 2 1 0PrefaceAlthough the principles of operation of helical screw machines, as compressors or expanders, have been well known for more than 100 years, it is only during the past 30 years thatthese machines have become widely used. The main reasons for the long period before they were adopted were their relatively poor efficiency and the high cost of manufacturing their rotors. Two main developments led to a solution to these difficulties. The first of these was the introduction of the asymmetric rotor profile in 1973. This reduced the blowhole area, which was the main source of internal leakage by approximately 90%, and thereby raised the thermodynamic efficiency of these machines, to roughly the same level as that of traditional reciprocating compressors. The second was the introduction of precise thread milling machine tools at approximately the same time. This made it possible to manufacture items of complex shape, such as the rotors, both accurately and cheaply.From then on, as a result of their ever improving efficiencies, high reliability and compact form, screw compressors have taken an increasing share of the compressor market, especially in the fields of compressed air production, and refrigeration and air conditioning, and today, a substantial proportion of compressors manufactured for industry are of this type.Despite, the now wide usage of screw compressors and the publication of many scientific papers on their development, only a handful of textbooks have been published to date, which give a rigorous exposition of the principles of their operation and none of these are in English.The publication of this volume coincides with the tenth anniversary of the establishment of the Centre for Positive Displacement Compressor Technology at City University, London, where much, if not all, of the material it contains was developed. Its aim is to give an up to date summary of the state of the art. Its availability in a single volume should then help engineers inindustry to replace design procedures based on the simple assumptions of the compression of a fixed mass of ideal gas, by more up to date methods. These are based on computer models, which simulate real compression and expansion processes more reliably, by allowing for leakage, inlet and outlet flow and other losses, VI Preface and the assumption of real fluid properties in the working process. Also, methods are given for developing rotor profiles, based on the mathematical theory of gearing, rather than empirical curve fitting. In addition, some description is included of procedures for the three dimensional modelling of heat and fluid flow through these machines and how interaction between the rotors and the casing produces performance changes, which hitherto could not be calculated. It is shown that only a relatively small number of input parameters is required to describe both the geometry and performance of screw compressors. This makes it easy to control the design process so that modifications can be cross referenced through design software programs, thus saving both computer resources and design time, when compared with traditional design procedures.All the analytical procedures described, have been tried and proven on machines currently in industrial production and have led to improvements in performance and reductions in size and cost, which were hardly considered possible ten years ago. Moreover, in all cases where these were applied, the improved accuracy of the analytical models has led to close agreement between predicted and measured performance which greatly reduced development time and cost. Additionally, the better understanding of the principles of operation brought about by such studies has led to an extension of the areas of application of screw compressors and expanders.It is hoped that this work will stimulate further interest in an area, where, though much progress has been made, significant advances are still possible.London, Nikola StosicFebruary 2005 Ian SmithAhmed KovacevicNotationA Area of passage cross section, oil droplet total surfacea Speed of soundC Rotor centre distance, specific heat capacity, turbulence model constantsd Oil droplet Sauter mean diametere Internal energyf Body forceh Specific enthalpy h = h(θ), convective heat transfer coefficient betweenoil and gasi Unit vectorI Unit tensork Conductivity, kinetic energy of turbulence, time constant m Massm˙ Inlet or exit mass flow rate m˙ = m˙ (θ)p Rotor lead, pressure in the working chamber p = p(θ)P Production of kinetic energy of turbulenceq Source term˙Q Heat transfer rate between the fluid and the compressor surroundin gs˙Q= ˙Q(θ)r Rotor radiuss Distance between the pole and rotor contact points, control volume surfacet TimeT Torque, Temperatureu Displacement of solidU Internal energyW Work outputv Velocityw Fluid velocityV Local volume of the compressor working chamber V = V (θ)˙VVolume flowVIII Notationx Rotor coordinate, dryness fraction, spatial coordinatey Rotor coordinatez Axial coordinateGreek Lettersα Temperature dilatation coefficientΓ Diffusion coefficientε Dissipation of kinetic energy of turbulenceηi Adiabatic efficiencyηt Isothermal efficiencyηv Volumetric efficiencySpecific variableφ Variableλ Lame coefficientμ Viscosityρ Densityσ Prand tl numberθ Rotor angle of rotationζ Compound, local and point resistance coefficientω Angular speed of rotationPrefixesd differentialΔ IncrementSubscriptseff Effectiveg Gasin Inflowf Saturated liquidg Saturated vapourind Indicatorl Leakageoil Oilout Outflowp Previous step in iterative calculations SolidT Turbulentw pitch circle1 main rotor, upstream condition2 gate rotor, downstream conditionContents1Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………………………. . . . . . . . . . . . . . . 1 1.1 Basic Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . 4 1.2 Types of Screw Compressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . ….. . . . . . . .7 1.2.1 The Oil Injected Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . …... . .71.2.2 The Oil Free Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . ….... .7 1.3 Screw Machine Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .8 1.4 Screw Compressor Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .101.5RecentDevelopments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.5.1RotorProfiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . 13 1.5.2CompressorDesign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2ScrewCompressorGeometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.1 The Envelope Method as a Basis for the Profiling of Screw CompressorRotors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………………………….. . . . . ….. . . . . . . . 19 2.2 Screw Compressor Rotor Profile s . . . . . . . . . . . . . . . . . . . . …. . . . . . . . . . . . . . . . . . . ….. . . 20 2.3 Rotor ProfileCalculation . . . . . . . . . . . . . . . . . . . . . . . . . . . …………………………. . . . . .23 2.4 Review of Most Popular Rotor Profiles . . . . . . . . . . . . . . . ………………………….. . . . . . 23 2.4.1 Demonstrator Rotor Profile (“N” Rotor Generated) . . ………………………………….. . 24 2.4.2 SKBK Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………... . . . . . . . . .26 2.4.3 Fu Sheng Profile . . . . . . . . . . . . . . . . . . . . . . . . . ………………………………. . . . . . . . .27 2.4.4 “Hyper”Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………………………………. . .27 2.4.5 “Sigma” Profile . . . . . . . . . . . . . . . . . . . . . . .. . . . . . ………………………………. . . . . .28 2.4.6 “Cyclon” Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………………………………. . . . . .28 2.4.7 Symmetric Prof ile . . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………… . . . . .29 2.4.8 SRM “A” Profile . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………… . . . . . . .30 2.4.9 SRM “D” Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………… . . . . . .31 2.4.10 SRM “G” Profile . . . . . . . . . . . . . . . .. . . . . . . . …………………………….. . . . . . . . . .32 2.4.11 City “N” Rack Generated Rotor Profile . . . . . . . . . . . ………………………………… . . 32 2.4.12 Characteristics of “N” Profile . . . . . . . . . . . . . . . . . . . ………………………………. . . . 34 2.4.13 Blower Rot or Profile . . . . . . . . . . . . . . . . . . . . …………………………….. . . . . . . . . . . 39 X Contents2.5 Identification of Rotor Positionin Compressor Bearings . . . . . . . . . . . . . . . . . . . . . . . . . . …………………………….. . . . . . . .40 2.6 Tools for Rotor Manufacture . . . . . . . . . . . . . . . . . . . . . . …………………………. . . . . . . .45 2.6.1 Hobbing Tools . . . . . . . . . . . . . . . . . . . . . . . . . . ………….…..………………. . . . . . . . . .45 2.6.2 Milling and Grinding Tools . . . . . . . . . . . . . . . . . . . ……………………………….... . . . . 482.6.3 Quantification of ManufacturingImperfections . . . . . ……………………………….... . . 483 Calculation of Screw Compressor Performance . . . . . . . . . . ………………………………. . . 49 3.1 One Dimensional Mathematical Model . . . . . . . . . . . . . . …………………………... . . . . . .49 3.1.1 Conservation Equationsfor Control Volume and Auxiliary Relationships . . . . ............................................... . . 50 3.1.2 Suction and Discharge Ports . . . . . . . . . . . . . . . . . . . ....................................... . . . . 53 3.1.3 Gas Leakages . . . . . . . . . . . . . . . . . . . . . . . . . . .................................... . . . . . . . . . .54 3.1.4 Oil or Liquid Injection . . . . . . . . . . . ...................................... . . . . . . . . . . . . . . . . . 55 3.1.5 Computation of Fluid Properties . . . . . . . . ........................................ . . . . . . . . . . . 57 3.1.6 Solution Procedure for Compressor Thermodynamics . (58)3.2 Compressor Integral Parameters . . . . . . . . . . . . . . . . . . . ………………………….. . . . . . . . 59 3.3 Pressure Forces Actingon Screw Compressor Rotors . . . . . . . . . . . . . . . . . . . . . . ................................... . . . . . . . 61 3.3.1 Calculation of Pressure Radial Forces and Torque . . . . .. (61)3.3.2 Rotor Bending Deflections . . . . . . . . . . . . . . . . . . . . . ……………………………….. . . . 64 3.4 Optimisation of the Screw Compressor Rotor Profile,Compressor Design and Operating Parameters . . . . . . . . . . ……………………………….. . . . 65 3.4.1 OptimisationRationale . . . . . . . . . . . . . . . . . . . . . . . . ……………………………….. . . . 65 3.4.2 Minimisation Method Usedin Screw CompressorOptimisation . . . . . . . . . . . ……………………………………… . . . . . . 67 3.5 Three Dimensional CFD and Structure Analysisof a Screw Compressor . . . . . . . . . . . . . . . . . . . . . . . . . …………………………….. . . . . . . . .71 4 Principles of Screw Compressor Design. . . . . . . . . . . …………………………… . . . . . . . . 77 4.1 Clearance Management. . . . . . . . . . . . . . . . . . . . . . . . ………….….………… . . . . . . . . . .78 4.1.1 Load Sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………….………………….. . . .79 4.1.2 Compressor Size and Scale . . . . . . . . . . . . . . ………………………………. . . . . . . . . . . 80 4.1.3 RotorConfiguration . . . . . . . . . . . . . . . . . . . . . . . ……………………………... . . . . . . .82 4.2 Calculation Example:5-6-128mm Oil-Flooded Air Compressor . . . . . . . . . . . . . . . ……………………………... . . . 824.2.1 Experimental Verification of the Model . . . . . . . . . . . ………………………………. . . . 845 Examples of Modern Screw Compressor Designs . . . . . . . ……………………………… . . . 89 5.1 Design of an Oil-Free Screw CompressorBased on 3-5 “N” Rotors . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………. . . . . . . 90 5.2 The Design of Familyof Oil-Flooded Screw Compressors Basedon 4-5 “N” Rotors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . …………………………… . . . . . . .93 Contents XI.5.3 Design of Replacement Rotorsfor Oil-FloodedCompressors . . . . . . . . . . . . . . . . . . . . . . . . . . . ................................. . .96 5.4 Design of Refrigeration Compressors . . . . . . . . . . . . . . . .............................. . . . . . . 100 5.4.1 Optimisation of Screw Compressors for Refrigeration . . . (102)5.4.2 Use of New Rotor Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . (103)5.4.3 Rotor Retrofits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ……………………………. . . 103 5.4.4 Motor Cooling Through the Superfeed Port in Semihermetic Compressors . . . . . . . . . . . . . . . . . . . …………………………………… . . . 103 5.4.5 Multirotor Screw Compressors . . . . . . . . . . . . . . . . . …………………………….... . . . . 104 5.5 Multifunctional Screw Machines . . . . . . . . . . . . . . . . . . ……………………….. . . . . . . . . 108 5.5.1 Simultaneous Compression and Expansionon One Pair of Rotors . . . . . . . . . . . . . . . . . . . . . . . . . . ............................................ . 108 5.5.2 Design Characteristics of Multifunctional Screw Rotors .. (109)5.5.3 Balancing Forces on Compressor-Expander Rotors . …………………..……………. . . 1105.5.4 Examples of Multifunctional Screw Machines . . . . . . . . (111)6Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . …………………… . . . . . . . . . 117A Envelope Method of Gearin g . . . . . . . . . . . . . . . . . . . . . . . . ………………………… . . . . . 119B Reynolds TransportTheorem. . . . . . . . . . . . . . . . . . . . . . . …………………………. . . . . . . 123C Estimation of Working Fluid Propertie s . . . . . . . . . . . . . . . …………………………….. . . . 127 Re ferences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ………………… . . . . . . . . . . 133中文译文螺杆压缩机N.斯托西奇史密斯先生A科瓦切维奇螺杆压缩机计算的数学模型和性能尼古拉教授斯托西奇教授伊恩史密斯博士艾哈迈德科瓦切维奇工程科学和数学北安普敦广场伦敦城市大学英国电子邮件:n.stosic@/doc/d6433edf534de518964bcf 84b9d528ea81c72f87.htmli.k.smith@/doc/d6433edf534de51896 4bcf84b9d528ea81c72f87.htmla.kovacevic@/doc/d6433edf534de51 8964bcf84b9d528ea81c72f87.html国会图书馆控制号:2004117305isbn-10 3-540-24275-9 纽约施普林格柏林海德堡isbn-13 978-3-540-24275-8 纽约施普林格柏林海德堡这项工作是受版权保护,我们保留所有权利。
压缩 机 中英词汇
Foundations 基础
Guide vane 导叶
Impeller 叶轮
Intercooling 中间冷却器
Coupling 联轴器
Diaphragm 隔板
Diffuser 扩压器
Discharge nozzle 排气接管
Discharge volute 排气蜗壳
Elastohydrodynamic 弹性流体动压
Elliptical 椭圆形
Film thickness 膜厚
Flow 流量
Fluid film 流体膜
Gas 气体
Hydrostatic 流体静压
Journal 轴颈
Liner 衬套
Materials 材料
Stress distribution 应力分布
Tie rods 拉杆
Damped systems 阻尼系统
Dewhirl vanes 破涡片
Inlet nozzle 进气(接)管
Inlet volute 进气蜗壳
Multistage 多级
Off design operation 非设计工况操作
Oil system 密封油系统
Absolute pressure 绝对压力
Absolute temperature 绝对温度
Adiabatic compression 绝热压缩过程
Air padding (压缩)空气(填充)输送
Aluminalkyle 烷基铝
Performance 性能
Slope 倾斜
Splitter vane 分流叶片
【机械类文献翻译】对螺杆压缩机的三维分析开发
英文原文Applications4.1 IntroductionThis chapter demonstrates the scope of the method developed for the three-dimensional analysis of a screw compressor. The CFD package used in this case was COMET developed by ICCM GmbH Hamburg, today a part of CD-Adapco. The analysis of the flow and performance characteristics of a number of types of screw machines is performed to demonstrate a variety of parameters used for grid generation and calculation.The first example is concerned with a dry air screw compressor. A common compressor casing is used with two alternative pairs of rotors. The rotors have identical overall geometric properties but different lobe profiles. The application of the adaptation technique enables convenient grid generation for geometrically different rotors. The results obtained by three dimensional modelling are compared with those derived from a one-dimensional model, previously verified by comparison with experimental data..The relative advantages of each rotor profile are demonstrated.The second example shows the application of three dimensional flow analysis to the simulation of an oil injected air compressor. The results, thus obtained, are compared with test results obtained by the authors from a compressor and test rig, designed and built at City University. They are presented in the form of both integral parameters and a p-∂indicator diagram. Calculations based on the assumptions of the laminar flow are compared to those of turbulent flow. The effect of grid size on the results is also considered and shown here.The third example gives the analysis of an oil injected compressor in an ammonia refrigeration plant.This utilises the real fluid property subroutines in the process calculations and demonstrates the blow hole area and the leakage flow through the compressor clearances.The fourth example presents two cases, one of a dry screw compressor to show the influence of thermal expansion of the rotor on screw compressor performance and one of a high pressure oil-flooded screw compressor to show the influence of high pressure loads upon the compressor performance.4.2 Flow in a Dry Screw CompressorDry screw compressors are commonly used to produce pressurised air, free of any oil. A typical example of such a machine, similar in configuration to the compressor modelled, is shown in Figure 4-1. This is a single stage machine with 4 male and 6 female rotor lobes. The male and female rotor outer diameters are 142.380 mm and 135.820 mm respectively, while their centre lines are 108.4 mm apart. The rotor length to main diameter ratio l/d=1.77. Thus, the rotor length∂=248.40 is driven at a speed of 6000 rpm by an is 252.0 mm. The male rotor with wrap angleωelectric motor through a gearbox. The male and female rotors are synchronised through timinggears with the same ratio as that of the compressor rotor lobes i.e. 1.5. The female rotor speed is therefore 4000 rpm. The male rotor tip speed is then 44.7m/s, which is a relatively low value for a dry air compressor. The working chamber is sealed from its bearings by a combination of lip and labyrinth seals.Each rotor is supported by one radial and one axial bearing, on the discharge end, and one radial bearing on the suction end of the compressor. The bearings are loaded by a high frequency force, which varies due to the pressure change within the working chamber. Both radial and axial forces, as well as the torque change with a frequency of 4 times the rotational speed. This corresponds to 400Hz and coincides with the number of working cycles that occur within the compressor per unit time.Figure 4-1 Cross section of a dry screw compressorThe compressor takes in air from the atmosphere and discharges it to a receiver at a constant output pressure of 3 bar. Although the pressure rise is moderate, leakage through radial gaps of 150 m is substantial. In many studies and modelling ,procedures, volumetric losses are assumed to be a linear function of the cross sectional area and the square root of pressure difference, assuming that the interlobe clearance is kept more or less constant by the synchronising gears. The leakage through the clearances is then proportional to the clearance gap and the length of the leakage line. However, a large clearance gap is needed to prevent contact with the housing caused by rotor deformation due to the pressure and temperature changes within the working chamber. Hence, the only way to reduce leakage is to minimise the length of the sealing line. This can be achieved by careful design of the screw rotor profile. Although minimising,leakage is an important means of improving a screw compressor efficiency, it is not the only one. Another is to increase the flow area between the lobes and thereby increase the compressor flow capacity, thereby reducing the relative effect of leakage. Modern profile generation methods take these various effects into account by means of optimisation procedures which lead to enlargement of the male rotor interlobes and reduction in the female rotor lobes. Thefemale rotor lobes are thereby strengthened and their deformation thus reduced.To demonstrate the improvements possible from rotor profile optimisation, a three dimensional flow analysis has been carried out for two different rotor profiles within the same compressor casing, as shown in Figure 4-2. Both rotors are of the “N” type and rack generated.Figure 4-2…N‟ Rotors, Case-1 upper, Case-2 lowerCase 1 is an o lder design, similar in shape to SRM “D” rotors. Its features imply that there is a large torque on the female rotor, the sealing line is relatively long and the female lobes are relatively weak.Case 2, shown on the bottom of Figure 4-2, has rotors optimised for operating on dry air. The female rotor is stronger and the male rotor is weaker. This results in higher delivery, a relatively shorter sealing line and less torque on the female rotor. All these features help to improve screw compressor performance.The results of these two analyses are presented in the form of velocity distributions in the planesdefined by cross-sections A-A and B-B, shown in Figure 4-1.In the case of this study, the effect of rotor profile changes on compressor integral performance parameters can be predicted fairly accurately with one-dimensional models, even if some of the detailed assumptions made in such analytical models are inaccurate. Hence the integral results obtained from the three-dimensional analysis are compared with those from a one-dimensional model.4.2.1 Grid Generation for a Dry Screw CompressorIn Case-1, the rotors are mapped with 52 numerical cells along the interlobe on the male rotor and 36 cells along each interlobe on the female rotor in the circumferential direction. This gives 208 and 216 numerical cells respectively in the circumferential direction for the male and female rotors. A total of 6 cells in the radial direction and 97 cells in the axial direction is specified for both rotors. This arrangement results in a numerical mesh with 327090 cells for the entire machine. The cross section for the Case-1 rotors is shown in Figure 4-3. The female rotor is relatively thin and has a large radius on the lobe tip. Therefore, it is more easily mapped than in Case-2 where the tip radius is smaller, as shown in Figure 4-4.Figure 4-3 Cross section through the numerical mesh for Case-1 rotorsThe rotors in Case 2 are mapped with 60 cells along the male rotor lobe and 40 cells along the female lobe, which gives 240 cells along both rotors in the circumferential direction. In the radial direction, the rotors are mapped with 6 cells while 111 cells are selected for mapping along the rotor axis. Thus, the entire working chamber for this compressor has 406570 cells. In this case, different mesh sizes are applied and different criteria are chosen for the boundary adaptation of these rotors. The main adaptation criterion selected for the rotors is the local radius curvature with a grid point ratio of 0.3 to obtain the desired quality of distribution along the rotor boundaries. By this means, the more curved rotors are mapped with only a slight increase in the grid size to obtain a reasonable value of the grid aspect ratio. To obtain a similar grid aspect ratiowithout adaptation, 85 cells would have been required instead of 60 along one interlobe on the female rotor. This would give 510 cells in the circumferential direction on each of rotors. If the number of cells in the radial direction is also increased to be 8 instead of 6 but the number of cells along axis is kept constant, the entire grid would contain more then a million cells which would, in turn, result in a significantly longer calculation time and an increased requirement for computer memory.Figure 4-4 Cross section through the numerical mesh for Case-2 rotors4.2.2 Mathematical Model for a Dry Screw CompressorThe mathematical model used is based on the momentum, energy and mass conservation equations as given in Chapter 2. The equation for space law conservation is calculated in the model in order to obtain cell face velocities caused by the mesh movement. The system of equations is closed by Stoke‟s, Fourier‟s and Fick‟s laws and the equation of state for an ideal gas. This defines all the properties needed for the solution of the governing equations.4.2.3 Comparison of the Two Different Rotor ProfilesThe results obtained for both Case 1 and Case 2 compressors are presented here. To establish the full range of working conditions and to obtain an increase of pressure from 1 to 3 bars between the compressor suction and discharge, 15 time steps were required. A further 25 time steps were then needed to complete the full compressor cycle. Each time step needed about 30 minutes running time on an 800 MHz AMD Athlon processor. The computer memory required was about 400 MB. In Figure 4-5 the velocity vectors in the cross and axial sections are compared. The top diagram is given for Case-1 rotors and the bottom one for Case-2. As may be seen, the Case 2 rotors realised a smoother velocity distribution than the Case 1 rotors. This may have some advantage and could have increased the compressor adiabatic efficiency by reduction in flow drag losses. In both cases, recirculation within the entrapped working chamber occurs as consequence of the drag forces in the air as shown in the figure. On the other hand, different fluid flow patterns can be observed inthe suction port. The velocities within the working chambers and the suction and discharge ports are kept relatively low while the flow through the clearance gaps changes rapidly and easily reaches sonic velocity.Figure 4-5 Velocity field in the compressor cross section for Case1 and Case2 rotorsFigure 4-6 Velocity field in the compressor axial section for Case1 and Case2 rotors These differences are confirmed in the view of the vertical compressor section through the female rotor axis, shown in Figure 4-6. In Case 2, lower velocities are achieved not only in the working chamber but also in the suction and discharge ports. In the suction port, this is significant because of the fluid recirculation which appears at the end of the port. This recirculation causes losses which cannot be recovered later in the compression process. Therefore, many compressors are designed with only an axial port instead of both, radial and axial ports. Such a situation reduces suction dynamic losses caused by recirculation but, on the other hand, increases thevelocity in the suction chamber which in turn decreases efficiency. Some of these problems can beavoided only by the design of screw compressor rotors with larger lobes and a bigger swept volume and a shape which allows the suction process to be completed more easily. However, rotor profile design based on existing one-dimensional procedures neglects flow variations in the ports and hence is inferior for this purpose. In such cases, only a full three dimensional approach such as this, will be effective.中文译文应用4.1简介本章介绍了对螺杆压缩机的三维分析开发的方法的范围。
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
英文原文Screw CompressorThe Symmetric profile has a huge blow-hole area which excludes it from any compressor applicat -ion where a high or even moderate pressure ratio is involved. However, the symmetric profile per -forms surprisingly well in low pressure compressor applications.More details about the circular p -rofile can be found in Margolis, 1978.2.4.8 SRM “A” ProfileThe SRM “A” profile is shown in Fig. 2.11. It retains all the favourable features of the symmetric profile like its simplicity while avoiding its main disadvantage,namely, the large blow-hole area. The main goal of reducing the blow hole area was achieved by allowing the tip points of the main and gate rotors to generate their counterparts, trochoids on the gate and main rotor respectively. T -he “A” profile consists mainly of circles on the gate rotor and one line which passes through the gate rotor axis.The set of primary curves consists of: D2C2, which is a circle on the gate rotor with the centre on the gate pitch circle, and C2B2, which is a circle on the gate rotor, the centre of whi ch lies outside the pitch circle region.This was a new feature which imposed some problems in the generation of its main rotor counterpart, because the mathematics used for profile generation at tha -t time was insufficient for general gearing. This eccentricity ensured that the pressure angles on th -e rotor pitches differ from zero, resulting in its ease of manufacture. Segment BA is a circle on th -e gate rotor with its centre on the pitch circle. The flat lobe sides on the main and gate rotors weregenerated as epi/hypocycloids by points G on the gate and H on the main rotor respectively. GF2 is a radial line at the gate rotor. This brought the same benefits to manufacturing as the previously mentioned circle eccentricity onFig. 2.11 SRM “A” Profile2.4 Review of Most Popular Rotor Profiles 31 the opposite lobe side. F2E2 is a circle with the cent -re on the gate pitch and finally, E2D2 is a circle with the centre on the gate axis.More details on t -he “A” profile are published by Amosov et al., 1977 and by Rinder, 1979.The “A” profile is a go od example of how a good and simple idea evolved into a complicated result. Thus the “A” pro file was continuously subjected to changes which resulted in the “C” profile. This was mainly gen erated to improve the profile manufacturability. Finally, a completely new profile, the“D” profile was generated to introduce a new development in profile gearing and to increase the gate rotor tor -que.Despite the complexity o f its final form the “A” profile emerged to be the most popular scre -w compressor profile, especially after its patent expired.2.4.9 SRM “D” ProfileThe SRM “D” profile, shown in Fig. 2.12, is generated exclusively by circles with the centres off the rotor pitch circles.Similar to the Demonstrator, C2D2 is an eccentric circle of radius r3 onthe gate rotor. B1C1 is an eccentric circle of radius r1, which, together withthe small circular arc A1J1 of radius r2, is positioned on the main rotor. G2H2is a small circular arc on the gate rotor and E2F2 is a circular arc on the gaterotor. F2G2 is a relatively large circular arc on the gate rotor which produces a corresponding curve of the smallest possible curvature on the main rotor.Both circular arc, B2C2 and F2G2 ensure a large radius of curvature in the pitch circle area. This avoids high stresses in the rotor contact region.Fig. 2.12 SRM “D” ProfileThe “G” profile was introduced by SRM in the late nineteen nineties as a replacement for the “D” rotor and is shown in Fig. 2.13. Compared with the“D” rotor, the “G” rotor has the unique feature of two additional circles in the addendum area on both lobes of the main rotor, close to the pitch circle.This feature improves the rotor contact and, additionally, generates shorter sealing lines. This can be seen in Fig. 2.13, where a rotor featuring “G” profile characteristics only on its flat side through segment H1I1 is presented.Fig. 2.13 SRM “G” Profile2.4.11 City “N” Rack Generated Rotor Profile“N” rotors are calculated by a rack generation procedure. This distinguishes them from any others. In this case, the large blow-hole area, which is a characteristic of rack generated rotors, is overcome by generating the high pressure side of the rack by means of a rotor conjugate procedure. This undercuts the single appropriate curve on the rack. Such a rack is then used for profiling both the main and the gate rotors. The method and its extensions were used by the authors to create a number of different rotor profiles, some of them used by Stosic et al., 1986, and Hanjalic and Stosic, 1994. One of the applications of the rack generation procedure is described in Stosic, 1996.The following is a brief description of a rack generated “N” rotor profile,typical of a family of rotor profiles designed for the efficient compression of air,common refrigerants and a number of process gases. The rotors are generated by the combined rack-rotor generation procedure whose features are such that it may be readily modified further to optimize performance for any specific application.2.4 Review of Most Popular Rotor Profiles 33The coordinates of all primary arcs on the rack are summarized here relative to the rack coordinate system. The lobe of the rack is divided into several arcs. The divisions between the profile arcs are denoted by capital letters and each arc is defined separately, as shown in the Figs.2.14 and 2.15 where the rack and the rotors are shown.Fig. 2.14 Rack generated “N” ProfileFig. 2.15 “N” rotor primary curves g iven on rack34 2 Screw Compressor GeometryAll curves are given as a “general arc” expressed as: axp + byq = 1. Thus straight lines, circles, parabolae, ellipses and hyperbolae are all easily described by selecting appropriate values for parameters a, b, p and q.Segment DE is a straight line on the rack, EF is a circular arc of radius r4,segment FG is a straight line for the upper involute, p = q = 1, while segment GH on the rack is a meshing curve generated by the circular arc G2H2 on the gate rotor. Segment HJ on the rack is a meshing curve generated by the circular arc H1J1 of radius r2 on the main rotor. Segment JA is a circular arc of radius r on the rack, AB is an arc which can be either a circle or a parabola, a hyperbola or an ellipse, segment BC is a straight line on the rack matching the involute on the rotor round lobe and CD is a circular arc on the rack, radius r3.More details of the “N” profile can be found in Stosic, 1994.2.4.12 Characteristics of “N” ProfileSample illustrations of the “N” profile in 2-3, 3-5, 4-5, 4-6, 5-6, 5-7 and 6-7 configurations are given in Figs. 2.16 to Fig. 2.23. It should be noted that all rotors considered were obtained automatically from a computer code by simply specifying the number of lobes in the main and gate rotors, and the lobe curves in the general form.A variety of modified profiles is possible. The “N” profile design is a compromise between full tightness, small blow-hole area, large displacement.Fig. 2.16 “N” Rotors in 2-3 configurationFig. 2.17 “N” Rotors in 3-5 configurationFig. 2.18 “N” Rotors in 4-5 configurationFig. 2.19 “N” Rotors in 4-6 configurationFig. 2.20 “N” Rotors compared with “Sigma”, SRM “D” and “Cyclon” rotorsFig. 2.21 “N” Rotors in 5-6 configurationFig. 2.22 “N” Rotors in 5-7 configurationFig. 2.23 “N” rotors in 6/7 configurationsealing lines, small confined volumes, involute rotor contact and proper gate rotor torque distribution together with high rotor mechanical rigidity.The number of lobes required varies according to the designated compressor duty. The 3/5 arrangement is most suited for dry air compression, the 4/5 and 5/6 for oil flooded compressors with a moderate pressure difference and the 6/7 for high pressure and large built-in volume ratio refrigeration applications.Although the full evaluation of a rotor profile requires more than just a geometric assessment, some of the key features of the “N” profile may be readily appreciated by comparing it with three of the most popular screw rotor profiles already described here, (a) The “Sigma” profile by Bammert,1979, (b) the SRM “D” profile by Astberg 1982, and (c) the “Cyclon” profile by Hough and Morris, 1984. All these rotors are shown in Fig. 2.20 where it can be seen that the “N” profiles have a grea ter throughput and a stiffer gate rotor for all cases when other characteristics such as the blow-hole area, confined volume and high pressure sealing line lengths are identical.Also, the low pressure sealing lines are shorter, but this is less important because the corresponding clearance can be kept small.The blow-hole area may be controlled by adjustment of the tip radii on both the main and gate rotors and also by making the gate outer diameter equal to or less than the pitch diameter. Also the sealing lines can be kept very short by constructing most of the rotor profile from circles whose centres are close to the pitch circle. But, any decrease in the blow-hole area will increasethe length of the sealing line on the flat rotor side. A compromise betweenthese trends is therefore required to obtain the best result.2.4 Review of Most Popular Rotor Profiles 39Rotor instability is often caused by the torque distribution in the gate rotor changing direction during a complete cycle. The profile generation procedure described in this paper makes itpossible to control the torque on the gate rotor and thus avoid such effects. Furthermore, full involute contact between the “N” rotors enables any additional contact load to be absorbed more easily than with any other type of rotor. Two rotor pairs are shown in Fig. 2.24 the first exhibits what is described as “negative” gate rotor torque while the second shows the more usual “positive” torque.Fig. 2.24 “N” with negative torque, left and positive torque, right2.4.13 Blower Rotor ProfileThe blower profile, shown in Fig. 2.25 is symmetrical. Therefore only one quarter of it needs to be specified in order to define the whole rotor. It consists of two segments, a very small circle on the rotor lobe tip and a straight line. The circle slides and generates cycloids, while the straight line generates involutes.Fig. 2.25 Blower profile中文译文螺杆压缩机螺杆压缩机的几何形状对称分布有一个巨大的吹孔面积不包括它任何压缩机应用在高或中等压力比参与。