Introduction Tilting pad journal bearings (TPJB) support rotating - PowerPoint PPT Presentation

May 2017 Year VI A CCOUNTING FOR T HERMALLY I NDUCED P AD D EFORMATIONS AND I MPROVING A F EEDING G ROOVE T HERMAL M IXING M ODEL TRC-B&C-01-17 Luis San Andrés Behzad Abdollahi Mast-Childs Professor Graduate research assistant A Computational Model for Tilting Pad Journal Bearings

Introduction Tilting pad journal bearings (TPJB) support rotating machinery with minimal destabilizing forces. Thermal effects Viscous shearing causes film y Lubricant in the temperature rise and power loss. sump Fluid film  High temperature in a pad (Babbitt liner becomes soft ~121 ° C).  Lubricant loses viscosity. Shaft rotation Pad  Hot clearance ↓ and preload ↑ speed, Ω to affect the x Static load, W minimum film thickness, load capacity, & Bearing bearing force coefficients. housing Pivot Aim : reduce temperature rise & Lubricant in the flow supplied with efficient Orifice groove lubricant delivery arrangements 2

Justification Introduction Goal: To accurately predict bearing performance without costly & time extensive tests. I. Improve a simple II. Account for temperature conventional model for induced deformation of pads, thermal and flow mixing in shaft, and housing affecting a groove region. bearing performance. 3

Temperature Field in Bearing Pads • Heat flows from film into shaft and pads. • 2D temperature distribution within a pad:         2 1 1 T T    0 r       2 2 r r r r • Heat convection boundary conditions on all sides of a pad. 4

Lubricant Mixing in a Feed Groove: Conventional Thermal mixing with hot oil carry over    Q Q Q sup LE TE   ( ) Q T Q T  sup sup TE LE T LE Q LE 𝝁 is an empirical coefficient • Required supply flow ( Q sup ) based only on upstream flow ( Q TE ) and downstream flow ( Q LE ). What if Q LE <  Q TE ?? • • During operation actual supply flow may differ from predicted (during design) • In practice, Q sup is controlled by available delivery system, rarely varying with operating condition. 5

Lubricant Mixing in a Feed Groove: He et al. Feed Groove supply flow rate Plenum 4 Q distributes Sup evenly as total Q Sup total 3 Q 4  i sup Q 4 Q sup n  TE 3 1 Q Q Sup Sup 1 W Q LE 1 2 3 4 Q Q Q Q 2 1 Sup Sup Sup Sup 2 Q Sup   1 4 1 ( ) Q Q Q Operation with journal eccentricity: LE TE sup • Pressure rise in a pad → demands of lesser supplied flow • Unloaded pads (low pressure) → demand too much lubricant He, M., Cloud, C. H., Byrne, J. M., and Vazquez, J. A., 2012, 41 st Turbomachinery Symposium 6 He, M., Allaire, P., Barrett, L., and Nicholas, J., 2005, Trib. Trans.

Novel Thermal Mixing Flow Model Flow at pad leading edge & trailing edge= shear flow -/+ pressure flow |   Q Q Q   , shear pressure LE TE    /2  L 3 R Lh h P  |    s   dz      , 2 12   LE TE R    s L /2 , LE TE Introduce a groove demand parameter i Q 1 |   ( C i ) to quantify the need of fresh flow shear C    1, , i i n i i Q Q from feed port: pressure TE • Shear driven flow is proportional to film thickness. ↑ demand • Film pressure gradient at leading edge induces a reverse flow. ↓ demand • Flow leaving upstream pad provides enough flow to fill in the downstream pad at its leading edge film. ↓ demand 7

Novel Thermal Mixing Model: Flow Distribution   n Total supply flow must meet total demand: C C total i  1 i A fraction of total supply flow is allocated to each feed port: C     i total total i ; 1, , Q Q Q i n sup sup i sup C total Example : flow fraction ( 𝜷 𝒋 ) for each feed groove vs. shaft speed & spec. load 4 pad bearing • Equal supply flow distribution at zero load (centered shaft). • As load increases, pad 1 and 2 require of lesser flow, while pad 3 receives most of the lubricant flow. • Pad 4 receives a large flow from upstream pad (3) → low demand of supply (fresh) flow. 8

Novel Thermal Mixing Model: Flow Distribution Total supply flow must meet total demand: C     i total total i ; 1, , Q Q Q i n sup sup i sup C total Example : flow fraction ( 𝜷 𝒋 ) for each feed groove vs. shaft speed & spec. load • Even supply flow distribution at 5 pad bearing zero load (centered shaft). • Pad 3 (loaded) receives a large flow from upstream pad → low demand • As shaft speed increases, the flow fraction amongst pads becomes more uniform. 9

Novel Thermal Mixing Model: Flow balance at a port • Excess supplied flow leaves a port as a side leakage Q flow. SL Q 2 Q TE   LE     1 i i i if Q Q Q TE sup LE (i-1) th pad     1 i i i i Q Q Q Q SL TE sup LE upstream Q i th pad gr • Additional required lubricant (downstream) is drawn from churning lubricant in a groove. Q Sup       1 i i i if Q Q Q TE sup LE Q SL     2 1 i i i i Q Q Q Q gr LE TE sup Based on descriptions in: Nicholas, J. C., Elliott, G., Shoup, T. P., and Martin, E., 2008, 37 th Turbomachinery Symposium 10 Ha, H. C., Kim, H. J., and Kim, K. W., 1995, ASME J. Tribol.

Novel Thermal Mixing Model: Energy Balance Thermal energy (heat) flows mainly by means of fluid motion,     advection heat transfer mechanism : c Q T p • Only inflowing streams (supply and trailing edge from upstream pad) carry in thermal energy. • Mixing efficiency parameter (0< C gr <1) Conservation of energy for sub control volumes: Side leakage stream ( Q SL , T SL ) Groove churning stream ( Q gr , T gr ) 11

Novel Thermal Mixing Model: Final Equation Control volume analysis: • Side leakage oil flow ( Q SL , T SL ) • Groove recirculating oil ( Q gr , T gr ) • Heat fluxes across the adjacent pad walls ( Φ LE , Φ TE ) From conservation of energy, film temperature at downstream pad leading edge:            1 1 i i i i i i i TE LE Q T Q T Q T Q T  TE TE sup sup SL SL gr gr c  i p T LE i Q LE 12

PREDICTIONS VS TEST DATA Model Validation 13

Case 1: A Five-Pad TPJB Hagemann et al. (2013) test a five pad bearing (ID 500 mm) for steam turbine. • Spray bars deliver fresh oil. • End baffle seals reduce required supply flow rate  flooded bearing. • Hollow shaft with pressure and displacement sensors is axially shifted. • Uses pressure and film thickness data to derive dynamic force coefficients. Hagemann et al., 2013, GT2013-95004; Kukla et al., 2013, GT2013-95074. 14

Case 1: Bearing geometry and operating conditions Shaft rotational speed Ω [RPM] 500 – 3000 C p /R = 0.0012 L/D = 0.7 Shaft surface speed Ω R [m/s] 13 – 79 1 – 2.5 Specific Load W/(LD) [MPa] TEHD predictions Load orientation LBP include both Number of pads 5 thermally and Shaft diameter [mm] 500 pressure induced Pad thickness [mm] 72.5 deformations Bearing axial length [mm] 350 Pad arc length 56° Pivot offset 0.6 Preload 0.23 Pad clearance [ μ m] 300 Lubricant ISO VG32 15

N = 3000 RPM Case 1: Pad Surface Temperature W/(LD) = 2.5 MPa Pad Surface Temperature Rise ( ° C) 75 New model New model improves prediction 60 over conventional C gr = 0.2 model predictions (up 45 Test data to 17 °C). 30 Conventional model Conventional model 15 predicts supply flow prediction, λ = 0.9 3 4 5 1 2 rate 70% larger than 0 -130 -30 70 170 actual. Angle (deg) • Flooded bearing → significant portion of lubricant leaving each pad (side or axially) is not immediately forced out. • Additional required oil to fill unloaded pads 1 & 5 is drawn from the captured oil inside the bearing housing. 16

N = 3000 RPM Case 1: Pad Surface Temperature W/(LD) = 1 MPa 75 Pad Surface Temperature Rise ( ° C) C gr = 0.2 Under two Test data, Q =210 L/min Prediction, Q =210 L/min 60 flow 45 conditions Test data, Q =420 L/min 30 Prediction, Q =420 L/min 15 3 5 1 2 4 0 -130 -30 70 170 Angle (deg) End-seal C/R= 0.004 Cut supply flow in half → ~ 20 ° C temperature raise. • Conventional model unable to account for operation with reduced flow. 17

Case 1: Pad Surface Temperature W/(LD) = 1 MPa Pad Surface Temperature Rise ( ° C) 1500 RPM 3000 RPM 4500 RPM Fix supply 100 1 5 2 3 4 80 New model, flow (420 Q =420 L/min C gr = 0.2 60 LPM) and 40 increase 20 shaft speed Simple model 0 -130 -30 70 170 Angle (deg) End-seal C/R= 0.004 Increase shaft speed by 1500 RPM → ~ 20 ° C temperature raise. 4500 RPM → Likely pad flow starvation. Conventional model predicts flow! 18

Case 2: A Four-Pad TPJB Coghlan & Childs (2014) test a four pad spherical seat TPJB with various oil feed arrangements and a constant (fixed) supply flow rate. » Flooded single-orifice (SO), labyrinth end seals » Evacuated leading edge groove (LEG) Evacuated spray-bar (SB) » Evacuated spray-bar blocker (SBB) » Coghlan, D. M., and Childs, D. W., 2015, GT2015-42331. 19 Coghlan, D. M., 2014, M.S. Thesis, Texas A&M University.