S TRUCTURAL FORCE COEFFICIENTS FROM METAL MESH PADS FOR A FOIL - - PowerPoint PPT Presentation

STRUCTURAL FORCE COEFFICIENTS FROM METAL

MESH PADS FOR A FOIL BEARING

Travis Cable

Graduate Research Assistant

Luis San Andrés

Mast-Childs Chair Professor

TRC-B&C-02-2016

TRC Project 400124-00078

May 2016 Year II

Justification

• Foil bearings are low maintenance mechanical

elements that dispense of expensive lubrication systems, saving on footprint, weight, and cost.

• Experiments with metal mesh dampers and small

metal mesh foil bearings show promising damping capabilities.

• OEMs/TM Users wish to extend metal mesh foil

bearings to large turbomachinery applications.

Objective and Tasks

Bring metal mesh foil bearings (radial & thrust) to a commercialization level

• 1. Construct jigs to manufacture metal mesh pads of various lengths: top foil

and bearing cartridge, along with a means to verify the pads and (assembled in) bearing static structural stiffness and material damping (loss factor).

• 2. Document manufacturing procedure and detail steps for

verification.

• 3. Build two low cost test rigs using commercial router

motors (25 krpm) to evaluate the static load performance and drag torque of radial and thrust MMFBs.

• 4. Measure rotor lift-off speed and break away torque, touchdown speed and

stall torque, load versus minimum film thickness, and drag power losses,

• ver a range of shaft speeds to 25 krpm.

Components of an Assembled Metal Mesh Bearing

Components:

• Bearing cartridge
• Stainless steel top foil
• Metal mesh underspring structure (pads)

Journal

Top foil

Metal mesh pad Bearing cartridge 90.17 mm

tMM

105.4 mm Bearing clearance tMM

Metal Mesh Pad Dimensions & Compactness Ratio

Five pads constructed (dimensions in mm)

• 7.36 mm thick pad gives a null clearance for the assembled bearing

Compactness ratio for a metal mesh pad:

MM copper MM

m CR V  

CR = 30%

CR ~ 30% is desirable

Industrial Metal Mesh

[a] Inconsistent wire mesh [b] Wire mesh from TWP Inc. [5]

Parameter Magnitude Mesh Size 16 per in Wire Diameter 0.011 in Opening 0.051 in Weight/square foot 0.14 lb/ft2 Density of copper 557 lb/ft3

Copper mesh from TWP Inc. Overlapping mesh makes for inconsistent thickness when compressed

Pad Forming

• 1. A length of copper mesh is cut

and weighed until achieving the desired mass (91.8 g) 1

• 2. The strip of mesh is then folded

end over end in a wooden jig to maintain the pads parallelism Digital Scale (+/- 0.05 g) 2

Pad Compression

A hydraulic press applies a load of 3,000 psi for a short time (10 sec.) to compress the metal mesh pad. After 10 s., pad removed from the compression jig and its thickness measured. If pad is thicker than desired, pad recompressed under a load of 500 psi until the desired thickness is obtained.

A Test Rig for Metal Mesh Pads

[a] Isometric view [b] Cross-section view

Test rig provides the ability to measure pad thickness, as well as perform static load vs deflection and dynamic load measurements.

Sliding Assembly

Static load tests: Sliding assembly connects to vertical mill chuck via a strain gauge load cell, moved up and down by mill lever.

Eddy current sensors Accelerometer location Load cell location Dowel pin locations

Dynamic load tests: Sliding assembly connects to an electromagnetic shaker via a dynamic load cell, moved up and down by the e-shaker permanent magnet.

[a] Isometric view [b] Side view

Thickness Measurements

MM test rig installed in a vertical mill. First touch thickness 2.22 N (0.5 lbf). Design = 7.36 mm First touch thickness measurements provide a repeatable and accurate metric for comparing metal mesh pads.

Metal Mesh Thickness [mm] Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Average Pad #1 7.59 7.59 7.59 7.59 7.59 7.59 Pad #2 7.54 7.54 7.54 7.54 7.54 7.54 Pad #3 7.61 7.61 7.61 7.61 7.61 7.61 Pad #4 7.29 7.29 7.29 7.29 7.29 7.29 Pad #5 7.60 7.60 7.60 7.60 7.60 7.60

Thickness measurements with calipers

• r

micrometers are inaccurate

Load vs. Deflection for MM Pads

• 1. Large initial displacements up to 100 N [W/Apad ~ 17 kPa)]
• 2. Pads demonstrate similar, but not identical structural
• stiffness. Pad #4 is noticeably different.
• 3. Max structural stiffness of ~ 2.9 MN/m for the largest

applied load of 900 N (W/Apad = 200 kPa)

[a] Load vs. deflection [b] Structural stiffness versus deflection

A Test Rig for Metal Mesh Pads

[a] Isometric view [b] SDOF Model

Measured dynamic force, sliding assembly acceleration and sliding assembly displacements. Modeled as a simple single degree of freedom (SDOF) system with a mass, linear spring and damper.

   

2

MM MM S

Mx C x K K x F t    

2

Re Im

eq MM

F M K X F C X                  

Dynamic Load Measurements for MM Pads

• 3. Trends and magnitudes are almost identical, despite

differences in pad thicknesses.

[a] Real part of complex stiffness [b] Imaginary part of complex stiffness

• 1. Real part of complex stiffness is representative of a SDOF

system with a linear spring

• 2. Imaginary part shows a general decrease with excitation

frequency, not representative of linear viscous damping.

Structural Stiffness for Metal Mesh Pads

Pad #1 Pad #2 Pad #3 Pad #4 Pad #5 Linear stiffness coefficient, KMM [MN/m] 0.96 1.70 1.64 1.96 1.59 Excited Mass, M [kg] 2.91 2.93 2.89 2.92 2.98 Natural Frequency, [Hz] 92 122 121 131 117 R2 Value [-] 0.986 0.994 0.995 0.996 0.995

MM

K M

• 1. Structural stiffness values for Pads 2-5 are similar

~ 1.75 MN/m.

• 2. Despite similar static load vs deflection results, Pad #1 is

displays lower dynamic stiffness KMM.

Applied mechanical preload W = 60 N, W/Apad = 14 kPa. Motion amplitude of 20 μm peak-peak

Material Loss Factor for Metal Mesh Pads

Material loss factors are similar (γ ~ 0.22). Uncertainty shows Pad #1 results are likely in error.

Applied mechanical preload W = 60 N, W/Apad = 14 kPa. Motion amplitude of 20 μm peak-peak

 

Im /

MM

F X K  

Material loss factors for Pads 1-5 Pad # 1 2 3 4 5 Average loss factor, γ 0.43 (±0.23) 0.22 (±0.05) 0.24 (±0.05) 0.20 (±0.04) 0.24 (±0.05)

Dynamic Load Measurements for Pad #5

[a] Increasing mechanical preload Fixed amplitude of motion of 20 μm peak-peak [b] Increasing amplitude of motion Fixed mechanical preload of W = 60 N, W/Apad = 14 kPa

• 1. Dynamic stiffness (KMM) for Pad #5 increases linearly with

mechanical preload, but decreases (linearly) with amplitude

• f motion.
• 2. Dynamic stiffness (KMM) for Pad #5 differs from the static

stiffness, derived from load vs. deflection tests. More pronounced at larger loads (W >30 N).

Dynamic Load Measurements for Pad #5

[a] Increasing mechanical preload Fixed amplitude of motion of 20 μm peak-peak [b] Increasing amplitude of motion Fixed mechanical preload of W = 60 N, W/Apad = 14 kPa

• 1. Loss factor (γ) for Pad #5 is not significantly affected by

mechanical preload or amplitude of motion.

• 2. General decrease of loss factor (γ) for Pad #5 with

increasing excitation frequency (30-300 Hz).

γavg ~ 0.22

Schematic of an Envisioned Test Rig

Router motor (40 krpm) Thrust runner and rotating shaft Aerostatic guide bearing Test thrust bearing holding apparatus Non-rotating section Torque measurement apparatus E-shaker Static loading plenum

Front View

Compression Spring Rotating section

Test rig similar to that presented in Refs. [6,7]

Loading Mechanism for a Test Rig

[a] Front view of a static loading mechanism for a foil bearing test rig [b] Cross section view of a static Loading mechanism for a foil bearing test rig

Test rig similar to that presented in Refs. [6,7]

Continuation Proposed Work 2016-2017

• 1. Refine test rig for the dynamic load characterization of

metal mesh pads.

• 2. Assemble a radial MMFB (5 pads), mount it atop a

rotor and measure its lift-off speed and other parameters for shaft speeds up to 40 krpm and an increasing static load (specific loads up to ~180 kPa, W ≤ 300 lbf).

• 3. Complete design and construct a novel metal mesh

foil bearing.

4.Complete the construction and troubleshoot a test rig for evaluation of foil thrust bearings and perform static load tests with the novel thrust bearing.

TRC Budget 2016-2017

Support for GS (20 h/week) x $ 2,400 x 12 months

$ 28,800

Fringe benefits (2.7%) and medical insurance ($377/month)

$ 5,300

Tuition three semesters ($ 363 credit hour x 24 ch/year)

$ 9,090

Travel and registration to a technical conference

$ 1,200 Total Cost:

$ 44,390

Questions (?)

TRC-B&C-02-16 STRUCTURAL FORCE COEFFICIENTS FROM METAL MESH PADS

FOR A FOIL BEARING

Travis Cable and Luis San Andrés

Thanks to TRC for their support

References

[1] Zarzour, M. and Vance, J., 2000, “Experimental Evaluation of a Metal Mesh Bearing Damper,” ASME J. Eng. Gas Turbines Power, 122, pp. 1-4. [2] Chirathadam, T. A., and San Andrés, L., 2012, “A Metal Mesh Foil Bearing and a Bump Type Foil Bearing: Comparison of Performance for Two Similar Size Gas Bearings,” ASME J. Eng. Gas Turbines Power, 134, pp. 10250. [3] De Santiago, O. and Solórzano, V., 2013, “Experiments with Scaled Foil Bearings in a Test Compressor Rotor,” Proc. ASME Turbo Expo, June, San Antonio, Texas, GT2013-94087, pp. 1-8. [4] Lee, Y.B., Kim, C.H., Kim, T.H. and Kim, T.Y., 2012, “Effects of Mesh Density

• n Static Load Performance of Metal Mesh Foil Bearings,” J. Eng. Gas Turbines

Power, vol. 134, pp. 1-8. [5] TWP Inc., 2014, “Copper Wire Mesh”, from http://www.twpinc.com/wire- mesh-material/copper. [6] Balducchi, F., Arghir, M., Gauthier, R. and Renard, E., 2013, “Experimental Analysis of the Start-Up Torque of a Mildly Loaded Foil Thrust Bearing,” ASME

• J. Trib., 135, pp. 1-7.

[7] Balducchi, F., Arghir, M. and Gauthier, R., 2015, “Experimental Analysis of the Dynamic Characteristics of a Foil Thrust Bearing ,” ASME J. Trib., 137, pp. 1-9.