May 2017 Year V EFFECT OF LUBRICANT SUPPLY PRESSURE ON SFD PERFORMANCE: ENDS SEALED WITH O-RINGS & PISTON RINGS TRC-SFD-01-17 Bonjin Koo Leping Yu Graduate Research Assistants Luis San Andrés Mast-Childs Chair Professor
Squeeze Film Damper (SFDs) Whirling motion from the journal squeezes the lubricant film and generates dynamic pressures that provides viscous damping to decrease rotor vibrations Aid to reduce rotor vibrations, suppress system instabilities, and provide mechanical isolation. Too little damping may not be enough to reduce vibrations. Too much damping may lock damper & will degrade system performance. 2
SFD Test Rig Structural stiffness (Rods), K S =1.6 MN/m, O-ring stiffness, K O-ring =0.6 MN/m, K S+O-ring =2.2MN/m Bearing cartridge mass, M BC =15 kg 3
Journal geometry and lubricant properties Short length SFD L/D=0.2 Piston ring seals Geometry Journal Diameter, D Land Length, L Radial Land Clearance, c Feed orifice Diameter, ϕ Feed orifice location End groove width End groove depth 127 mm (5.0 in) 25.4 mm (1.0 in) 373 μm (14.7 mil) 2.54 mm (0.1 inch) 45º 2.5 mm 3.8 mm Supply orifices Lubricant Properties Lubricant Type ISO VG 2 Supply temperature, T in 23 C (73 F) Lubricant viscosity @ T in, μ 2.57 cp Lubricant density, ρ 820 kg/m 3 4
PR-SFD: Piston rings as end seals (θ=135 o ) Piston ring (PR) Bearing Cartridge Journal PR slit Y θ=90 Static Loader θ=45 θ=0 X Feedhole Lubricant Leakage Journal Film Housing Piston rings (a) Piston ring Outer diameter: 127 mm Thickness: 3.3 mm Piston ring geometry Outer diameter 127 mm Thickness 3.3 mm Width 2.5 mm Material Steel 5
OR-SFD: O-rings as end seals Journal Y θ=90 No leakage O-ring Feedhole (φ=2.3 mm) Journal Housing Discharge hole (φ=2.0 mm) θ=0 X Bearing Cartridge Lubricant Discharge Film (b) O-ring Steel spacer O-rings Diameter: 120 mm Thickness: 2.6 mm O-ring geometry Diameter 120 mm Thickness 2.6 mm Material Buna-N 6
Test Procedure Step 1: Apply loads and measure BC motions Apply forces by shakers F 1 1 F X iωt = Re e 1 if Y CW Y X F 2 2 F X iωt = Re e 2 if Y CCW Y X Record BC displacement z and acceleration a 1 z 1 1 x (t) X 1 1 e iωt = = y (t) Y 1 a x X e i t 2 2 2 (t) ω 2 z = = 2 2 a y(t) Y EOM: Frequency domain 2 [ L + iω L ω L] = MBC K C M z F a Find parameters: H = K ω M + iωc L 2 L L L 7
Test Procedure Step 2: Curve fit H L s using KCM model c=373 µm, orbit size r/c=0.3 (K, C, M) SFD = (K, C,M) L (K, C, M) S ( M BC ) SFD Film Re [ F az ] K ω M 1 2 L L Test system (Lubricated) Dry structure 1 ( M BC ) Im [ F az ] C L ω P s =1.4 bar Test data H XX H YY Model fit P s =3.5 bar Test data Model fit Test data Model fit H XX H YY P s =6.2 bar H XX H YY 8
OR-SFD damping and mass vs. supply pressure Journal Y θ=90 O-ring Discharge hole (φ=2.0 mm) θ=0 X Bearing Cartridge Feedhole (φ=2.3 mm) c=373 µm, r/c=0.3, ω=10-100 Hz P s >2.0 bar, C avg and M avg remain at ~ 11kN-s/m and ~30 kg, respectively. P s <2.0 bar, C avg as P s. 9
OR-SFD & PR-SFD C and M vs. supply pressure c=373 µm, r/c=0.3, ω=10-100 Hz - Damping C avg for OR-SFD is 11% larger than C avg for PR-SFD. - Added mass M avg ~30 kg as supply pressure decreases. - Damping coefficients C avg as lubricant supply pressure. 10
Peak film pressure vs. freq Peak-peak dynamic pressure [bar] PR-SFD r/c=0.3 P s =0.7 bar P s =3.5 bar P s =6.2 bar Air ingestion Peak-peak dynamic pressure [bar] c=373 µm, r/c=0.3 OR-SFD r/c=0.3 P s =0.7 bar P s =3.5 bar P s =6.2 bar Air ingestion Frequency [Hz] Frequency [Hz] For P s =0.7 bar, and whirl freq. > 60 Hz, peak-to-peak dynamic pressure stops growing (due to air ingestion). 11
PR-SFD pressure profiles P s =0.7 bar P s =2.1 bar P s =3.5 bar r/c=0.3, ω= 90 Hz Pressure profiles for P s =3.5 and 6.2 bar are almost identical. Spikes in pressure may be due to bursts of leakage thru PR slits. P s =6.2 bar 12
OR-SFD pressure profiles P s =0.7 bar P s =2.1 bar r/c=0.3, ω= 90 Hz Pressure profiles for P s =3.5 and 6.2 bar are ~ identical. There are no sharp spikes on pressure. P s =3.5 bar Y θ=90 O-ring θ=0 X P s =6.2 bar P 4 (θ=225 o ) Measured point Discharge hole (φ=2.0 mm) 13
Sudden loss of flow Tests conducted with SFD operating (under dynamic load) with amplitude r=0.2c. Pressure supply (flow rate) is set. At time t=0 s, flow (pressure) is cut off. Ensuing motions recorded as squeeze action expels remnant lubricant in film. 14
Test rig time response P s =Q s =0.0 at t >0 P s =3.5 bar, Q s =2.4 LPM P s =6.2 bar, Q s =3.0 LPM PR-SFD Clearance Clearance PR slit (θ=135 o ) Y θ=90 Feedhole (θ=45 o ) θ=0 X Displacement, Y (µm) Out of sensor range Displacement, Y (µm) Out of sensor range Displacement, X (µm) Displacement, X (µm) OR-SFD P s =3.5 bar, Q s =3.5 LPM P s =6.2 bar, Q s =6.0 LPM Discharge hole (φ=2.0 mm) Y θ=90 Feedhole (φ=2.3 mm) θ=0 X O-ring Displacement, Y (µm) Clearance Recorded time 0 sec 1 sec 2 sec 3 sec 4 sec 5 sec Jump & static offset Displacement, X (µm) 15
WATERFALLs of motion PR-SFD P s =Q s =0.0 at t >0 P s =3.5 bar, Q s =2.4 LPM PR slit (θ=135 o ) Y θ=90 Feedhole (θ=45 o ) θ=0 X OR-SFD P s =3.5 bar, Q s =4.5 LPM Jump & static offset Y θ=90 Feedhole (φ=2.3 mm) Discharge hole (φ=2.0 mm) θ=0 X O-ring 16
Conclusion Both O-ring and piston-ring sealed ends SFDs show similar damping and added mass coefficients. For a large supply pressure (P s ) performance of a sealed ends SFD does not change with an increase in squeeze film velocity (V s =rω). Too low oil feed pressure reduces film pressure and damping. A sudden loss of flow causes immediate changes in performance: (a) For PR-SFD, whirl orbit motion increases in amplitude (w/o bound) to show collapse of element (to touch clearance). (b) For OR-SFD, whirl motion at t=0 jumps (static offset) and later shows growth. O-ring resilience keeps motions bounded 17
Acknowledgments Turbomachinery Research Consortium & Pratt & Whitney Engines Questions (?)