Disc Bearing Specification Review and Potential Revision

Dr. Sougata Roy

AASHTO COBS Meeting, Montgomery, Alabama, 2019T-2: AASHTO Technical Committee for Bearings and Expansion DevicesAlabama C, June 24, 8:00 – 12:00 Noon

Update on Data Gathering and Potential Revision

Specifications for Disc Bearings

• High Load Multi Rotational (HLMR) Bearings

• Provisions of Disc Bearings are not adequately developed

• Relevant Specifications• AASHTO LRFD Bridge Design Specifications (BDS)

• AASHTO LRFD Construction Specifications (BCS)

• AASHTO Material Specifications

• ASTM Specifications

Objective

• Develop/Elaborate the design, construction and material specification for Disc Bearings

• Review all relevant specifications and propose modifications to rectify any inconsistencies and other shortcomings

Tasks

• Review specifications

• Literature search

• Review industry documents/proposal

• Identify areas requiring change

• Develop proposals for change

• Discuss with T-2

• Discuss with industry

• Incorporate changes

Goal: Potential ballot items at the COBS Meeting, 2019

Progress

• Reviewed AASHTO LRFD BDS

• Reviewed AASHTO LRFD BCS

• Reviewed/Reviewing othe relevant specifications

• Reviewed/Reviewing published literature

• Reviewing comments/proposals received from industry

Specifications• AASHTO LRFD BDS Article 14.7.8

• AASHTO LRFD BCS Section 18, Relevant Articles

• AASHTO M251-06 – Elastomeric Bearings

• ASTM D4014 – Elastomeric Bearings

• ASTM D395 –Compression Set

• ASTM D412 –Tension

• ASTM D429 –Adhesion to Substrate

• ASTM D746 –Brittleness Temperature

• ASTM D2240 –Durometer Hardness

• ASTM D2990 – Creep

• ASTM D6147 – Stress Relaxation

• Others…

Material (Elastomeric Disc)

• Polyehter urethane• Stiff for vertical deflection

• Flexible for rotation without liftoff

• Creep should be taken into account

• Virgin material• Hardness 45-65 Shore D

PINBOTTOM PLATE

T O P P L A T E

AASHTO LRFD BDS - Elastomeric Disc• Service Limit State

• Δ𝑖 ≤ 0.1ℎ

• Δ𝑐𝑟 ≤ 0.08ℎ

• No lift off

• 𝜎𝑐,𝑎𝑣𝑔 ≤ 5.0 𝑘𝑠𝑖

• Strength Limit State• Design for 𝛿𝑢, 𝜃𝑢

AASHTO LRFD BCS

AASHTO LRFD BCS

• Elaborate on PU disc Fabrication?• Surface finish of PU disc?• Clarification on surface coating?

Fabrication Tolerance

QC and QA

• 18.3.4.2 Material Certification and Performance Testing by Manufacturer

• Minimum 2 bearings

• Minimum 2 individual bearing components

• 18.3.4.3 Quality Assurance Testing by the Engineer• Only if required by the contract (random sampling)

• At least one set of material property tests per lot

18.3.4.4.1 Material Certification Testing

Dimension Check and Clearance Test

• Gap between disc and retaining pin?• Gap between pin and plate?

Long Term Deterioration Test

• Elaborate acceptance criteria?

Example

Review Comments of RJW Proposal – Stress Limits

• First develop specifications for 𝜎𝑐 ≤ 5 𝑘𝑠𝑖

• Concern about uniformity of material production and performance

• Concern about long term performance and creep deformation

Hardness (Durometer)

RJW

AASHTO

• Too relaxed hardness limits• Retain AASHTO limits

https://gallaghercorp.com/thermoset-urethane-vs-thermoplastic-urethane/

Modulus of Elasticity – 𝐸 and 𝐸𝑐

• 𝐸: Actual modulus of elasticity • can be obtained from stress-strain response of material

• 𝐸𝑐: Apparent modulus of elasticity • Including the effect of shape factor, S, and boundary

conditions of the loaded surface

• 𝐸𝑐 = 𝐸(𝑎 + 𝑏𝑆2)• 𝑎, 𝑏 to be determined from experiments and depends

on boundary conditions (friction) of the loaded surface

Recommended 𝐸𝑐 (RJW)

• 𝐸𝑐 = 𝐸 1 + 𝑆2

• Assumes 𝑎 = 1, 𝑏 = 1

• Only valid for high friction and practically no slip at the loaded surface

• Not always the case

• Test data did not correlate well due to significant nonlinear stiffening response

• Slip reduces apparent stiffness, results increased deflection

• Concerns with retainer pin touching the base plate and affecting bearing performance

Test Result

Disc Design Method B, PBEng, Dr. Paul Bradford

Friction

• Not measured in RJW tests

• Important for performance

• Needs to be verified for conformance with design by proof testing

• It would be desirable also to check Poisson Ratio• Probably close to 0.48

Low Friction – Axial Compression

High Friction – Axial Compression

Low Friction - Shear

High Friction - Shear

Shear Stress at Top Plate Interface

Axial Stress at Top Plate Interface

Surface Finish and Friction

Shape Factor, S

• 𝑆 =𝐿𝑜𝑎𝑑𝑒𝑑 𝐴𝑟𝑒𝑎

𝐴𝑟𝑒𝑎 𝐹𝑟𝑒𝑒 𝑡𝑜 𝐵𝑢𝑙𝑔𝑒

• For a circular disc

𝑆 =

𝜋𝐷2

4𝜋𝐷ℎ

=𝐷

4ℎ

• Thinner disc, i.e., larger 𝐷

ℎ=> larger stiffness and

larger stress carrying capacity (smaller strain)

• Thicker disc, i.e., smaller 𝐷

ℎ=> smaller stiffness and

larger strain (smaller stress)

Annular Disc (Donut Shape)• If inside free to bulge

𝑆 =

𝜋4(𝐷𝑜

2 − 𝐷𝑖2)

𝜋 𝐷𝑜 + 𝐷𝑖 ℎ=𝐷𝑜4ℎ

1 −𝐷𝑖𝐷𝑜

= 𝑆𝑜(1 − 𝛽)

• If inside is constrained

𝑆 =

𝜋4(𝐷𝑜

2 − 𝐷𝑖2)

𝜋𝐷𝑜ℎ=𝐷𝑜4ℎ

1 −𝐷𝑖2

𝐷𝑜2 = 𝑆𝑜(1 − 𝛽2)

• For Disc Bearings, 0.2 ≤ 𝛽 ≤ 0.5• If inside free to bulge: 0.5𝑆𝑜 ≤ 𝑆 ≤ 0.8𝑆𝑜• If inside constrained : 0.75So ≤ S ≤ 0.96𝑆𝑜

• For Disc Bearings, 𝑆≈2

Suggested Design Specification

• Based on existing design practice (RJW)• 𝜎𝑐 ≤ 5.0 𝑘𝑠𝑖

• 𝜖𝑐 ≤ 0.1

• 𝛼𝑚𝑖𝑛 ≥ 0.013 𝑟𝑎𝑑 (scaled from 0.02 radians at 150% of the specified rated capacity for proof load test)

• Limit on 𝜎𝑐𝜖𝑐?

• Shear strain limit?

Shear Strain – Circular Disc

𝛾𝑐~6𝑆𝜖𝑐 − 36𝑆3𝐺

𝐾𝜖𝑐

For small S (<4)𝛾𝑐~6𝑆𝜖𝑐

Constantinou et al. (1992)

Shear Strain – Annular Disc

Constantinou et al. (1992)

Compression Modulus

Constantinou et al. (1992)

Clearances

• Between PU disc and retaining pin

• Between plate and retaining pin

Required Specification Changes• Material properties of Polyether Urethane

• Clarification of PU (elastomer?)• Specification of Polyether Urethane – ASTM?

• Shear strength and modulus• Creep • Low temperature performance

• Fabrication• Casting (what about extrusion?)

• Any special requirements or QC?

• Surface Finish for PU

• Clarification on surface coating for Steel Plates

• Clarification on acceptance criteria after surface deterioration test

• Testing for• Shear strength and modulus• Creep• Friction between PU and Steel Plate

Groove Shape

• Angular – 45, 60 degrees?

• Parabolic?

Additional Suggestions, Comments, Questions, Criticisms?

Input from all stake holders are welcome

Request for InformationTo all,

You are receiving this because you are known to be or may be a producer of disc bearings. The AASHTO T-2 Technical Committee on Bearings and Expansion Devices is reviewing AASHTO specifications for design and construction of disc bearings, to identify any potentially needed updates, prior to a longer term effort that may include increases in the allowable design loads on this type of bearing. The T-2 Committee is seeking voluntarily provided data from producers of these bearings, for confirmation of acceptability of existing specifications and to identify areas that may need improvement. If you manufacture this type of bearing, we are requesting all or part of the following information that you are willing to provide:

(1) Specific fabrication specifications for Polyurethane (PU) discs, including surface finish. Please include any particular ASTM specifications used for procuring the PU material.

(2) Physical properties of the PU as per Table 18.3.2.8-1 of the AASHTO Bridge Construction Specifications. Please include the specimen sizes and reference codes for these tests.

(3) Creep performance data including specimen size and reference codes.

(4) Any measurement of Poisson ratio (or bulk modulus).

(5) Coefficient of friction between the PU disc and the top and bottom plates (as required by Table 18.3.4.3-1).

(6) Measurements of modulus of elasticity (stress-strain curve) and shear modulus (shear stress - shear strain curve) and any such values used for design.

(7) QC data pertaining to dimensional checks and clearance test (such as the gap between the shear pin and the inner face of the disc and the bottom plate).

(8) Long-term deterioration test results including documented surface condition after testing and measured coefficient of friction (as required by the specs.). Examples of rejected specimens would be particularly helpful in establishing a suitable acceptance criterion.

(9) Any data on low temperature performance.

(10) Any specific design concerns.

Providing all or part of this information is completely voluntary. If you are willing to provide information but would like theinformation to be anonymous, you can e-mail the information to me as an attachment that does not include your company’s name. I will then only share the information as anonymous information when sharing it with other T-2 Committee members and with the consultant assisting the T-2 Committee. If possible, please provide the information by January 25th, 2019. If you are willing to provide information, but a different date for providing the information is needed, please let me know. Thankyou.

Carl Puzey, P.E., S.E.Illinois Department of TransportationEngineer of Bridges & Structures

Specific fabrication specifications for Polyurethane (PU) discs, including surface finish.

Any particular ASTM specifications used for procuring the PU material

• Response 1roughness of the urethane disc is not as important as the roughness of the plate it is mating with, and not necessary in the specifications. Average roughness of the top and bottom urethane surfaces around 450 micro-in.

• Response 2See AASHTO Table 18.3.2.8-1: Physical Properties of Polyether Urethane.

Physical properties of the PU as per Table 18.3.2.8-1 of the AASHTO Bridge Construction Specifications. Include the specimen sizes and reference codes for these tests.

• Response 1ASTM test methods as shown in the AASHTO table. The ASTM test specifications would also designate specimen sizes, which will vary by test. Results are confirmed by urethane suppliers, and sometimes also validated by owner labs, when required.

• Response 2See Urethane Supplier Handbook.

Creep performance data including specimen size and reference codes.

• Response 1This is important and highly influenced by roughness of plate mating with urethane. Does not have data on creep, and is not aware of an applicable ASTM test method.

• Response 2See Urethane Supplier Handbook.

Any measurement of Poisson ratio (or bulk modulus).

• Response 1This value is not provided by urethane supplier.

• Response 2Not measured, but estimated to be approximately 0.485.

Coefficient of friction between the PU disc and the top and bottom plates (as required by Table 18.3.4.3-1)

• Response 1The coefficient of friction test cited in AASHTO Table 18.3.4.3-1 pertains to PTFE sliding coefficient of friction. Metallized contact surface with PU provides sufficient roughness to limit creep.

• Response 2Not measured, but estimated to be 0.8.

Measurements of modulus of elasticity (stress-strain curve) and shear modulus (shear stress - shear strain curve) and any such values used for design

• Response 1Stress strain values vary based on shape factor, similar to other elastomers. Compression-deflection values per ASTM D575 provided by PU supplier. However, these values are not directly used for disc designs, since shape factor needs to be considered.

• Response 2Using empirical equation including shape factor.

QC data pertaining to dimensional checks and clearance test (such as the gap between the shear pin and the inner face of the disc and the bottom plate)

• Response 11/16” oversize (1/32” design clearance) as a standard; increased dimension based on geometric check if needed based on pin size and rotation to avoid binding. This check is roughly equivalent to the AASHTO LRFD pot-piston clearance check in 14.7.4.7-5.

Long-term deterioration test results including documented surface condition after testing and measured coefficient of friction (as required by the specs.). Examples of rejected specimens would be particularly helpful in establishing a suitable acceptance criterion.

• Response 1Example provided. Wear is observed due to extreme tests, but does not impact performance.

• Response 2No specimen has been rejected in long-term testing.

Any data on low temperature performance.

• Response 1No

• Response 2See PU Supplier Handbook

Any specific design concern.

• Address importance of plate roughness in contact with urethane disc (better than limiting ring). Something be added to specifications.

• Address requirements for bearing replaceability. Several ways to achieve: Additional bearing plates that are bolted or recessed into sole/masonry plates, use of coupler nut system for bearing anchorage to substructure, etc.

• Address shear restricting pin connection type (threaded, welded, press fit) and design guidelines. Threaded pins might perform considerably better than the other two options. Assuming threaded pins are required, address pin thread specifications (depth and pitch).

• Address required pin orientation - Is attaching the pin to upper or lower plates acceptable? Upper part of assembly rotates, so inverted pins attached to the upper plate sometimes gouges the lower plate when the bearing is rotated, or worse, rotated and compressed. Currently suppliers can design either configuration, which may or may not be an issue but should be mentioned.

• Address pin thread engagement into the bearing plate. The pin is not in full bearing, and when not threaded deep enough will lean under horizontal loading (failure of pin thread/pin hole essentially due to overstress at that interface). Design requirements against bending/overturning.

Any specific design concern.

• Address bearing design requirements for uplift under service loads and strength/extreme combinations. Fatigue design requirements. Provide guidelines/equations various load cases.

• Design specifications should be provided in AASHTO similar to the pot bearing section; otherwise there are safety concerns with each manufacturer designing differently, and/or new manufacturers entering the market without experience and under designing components or interfaces.

• Design thickness of disc should be based on rotation and diameter

• Clearly specify requirements for roughness and corrosion protection of plates in contact with disc.

• PU material is over-designed. Permit higher allowable stress.

• If allowable stress is increased, address what other low friction (higher allowable stress) materials can be allowed for main sliding surfaces besides PTFE.

Any specific design concern.

• In the latest AASHTO LRFD Specifications very little is mentioned about the use of Bottom (Lower) steel Bearing Plates. Only in section 14.7.8.5 – Steel Plates, is there any mention of these plates and it only gives reference on the minimum thickness that is required. It may be in the Committee’s best interest to at least suggest that lower bearing plates be mandated (as many State DOT’s have already done in their specifications), for bearing replacement that requires very small jacking height limits as stated in 14.8.1. The elimination of these lower bearing plates requires that the jacking height must be sufficient to clear the shear pin from its hole in the masonry plate. This usually requires considerable jacking height to achieve this clearance.

Other Comments Received

• According to analyses an effective coefficient of friction of about 0.80 is required for good disc performance. The actual value of friction is not important, the interfaces should be rough/grooved, or a retaining ring/recess should be used. Practically speaking, if, based on test results the disc slips too much, then the surface needs to be roughened. Slip is not necessarily a bad thing, it relieves surface shear stresses near the edges. Discs are grooved in part to promote that behavior. Excessive slip is what the manufacturer needs to prevent.

Other Comments Received

• Disc scuffing on the long term deterioration test is typical, it shows that that friction is working to resist slip. Long term deterioration is a misnomer for that test, a disc will support a structure for 100 years and look nothing like test specimens. The test spec needs to be tweaked to better reflect field conditions, and categorized as a certification rather than a project QC test. Dr. Stanton, who ran the research program incorporating this test, has made many presentations over the years related to the conservatism of the test and disc durability.

Other Comments Received

• Material ingress into the pin/ring region is not a problematic detail. Proof load tests require rotation and inspection. If the inspection reveals that the disc is being pinched, the manufacturer should adjust their design accordingly. Material pinching on the ID has never caused a performance problem.

Other Comments Received

• Past experience and analysis has shown that 30 degree from perpendicular edge grooves work best. The idea of elliptical grooves has been investigated many times over the past few decades. The conclusion always seems to come down to that discs don’t fail with grooves that are sharper then elliptical, so an elliptical groove is seen as a nonvalue added detail.

• While the use of a groove is beneficial, the tact of AASHTO specifying the shape of the groove seems heavy handed. In my opinion the appropriate level of regulation is for AASHTO to encourage the use of grooves. This can be done through commentary and removing the penalty that currently exist for including grooves. Grooves are an added cost, not only in terms of machining/casting the groove, but they increase the overall size of the bearing. Currently AASHTO places a 5 ksi compressive stress limit on the disc based on the ungrooved area of the disc. Thus a grooved disc has to be a little bit larger than an ungrooved disc, despite the fact that grooved discs perform better than ungrooved discs (reduces slip and shear strain at the disc OD). From an economic perspective, AASHTO implicitly discourages the use of grooves.

• Suggest limiting the groove to 45 deg and making the disc OD as the basis for average compressive stress.

Other Comments Received

• Material shear tests, such as an ASTM D4014 quad shear test, would provide interesting material information, but for what purpose? Shear strain due to lateral movement is restricted by the shear restriction mechanism. There will be some level of correlation between durometer and shear modulus, but exactly how that maps to design is an open question. Perhaps as a material characterization test for some sort of reoccurring certification process it would make sense (e.g. once every 2-3 years), but to require for every project would add cost with little or no value.

Other Comments Received

• A disc is pretty rugged component, it can support load with cracks and holes and overload. Currently stress relaxation testing, established by elastomeric bearing industry, is used to identify material compounds that may display excessive creep, and should be adequate for PU discs.

• Creep is related to the bearing design. Perhaps as part of a certification process a sustained load test on a full-sized bearing would be of value.

• Not sure the current creep specification limit of 0.08h is adequate. A 1” thick disc bearing that creeps 0.08” is probably of no concern, but 0.08h creep may be for a 4” thick disc. This specification item could be improved.

• Adding in a small amount of vibration would help promote any creeping action that might take place.

• Creep takes place on a logarithmic time scale, i.e. most of it takes place early on. Sustained load test of 3 days to a week should be adequate for identifying any fundamental problems..

Other Comments Received

• Direct friction testing is unnecessary and may be cumbersome to implement. Compression-deflection bearing testing will identify whether the surface friction is adequate or not.

RJW Presentation at IBC 2019

• Provided overview of Non-destructive Testing of HLMR Bearings (significant focus on disc bearings)

• Provided background to long term deterioration testing

• Disc bearing shows cosmetic damage under long term deterioration testing that does not compromise the disc performance

• This test should only be considered periodically (not on each project) as a qualification testing of supplier

• Testing of a bearing that was in service for more than 30 years(?) showed no degradation in performance

Possible Specification Improvements – Short Term

• Four plate design (sole, top, bottom, masonary)

• Material Testing for Creep

• QA/QC – verification of surface friction • Coefficient of Friction of at least 0.8?

• Corrosion protection of mating faces of top and bottom plates by metallizing

• Paint to be prohibited

• Dimension Check and Clearance Test• Gap between retaining pin and plates?

Possible Specification Improvements – Long Term• Design Provisions

• S (pin clearance)• Ec (requires knowledge of G) • sc, gc

• Creep deformation (pin clearance)• Rotation, thickness • Uplift• Lateral loading (pin design)

• Long Term Deterioration Testing• Requirement (Frequency)• Acceptance Criterion

• PU Specifications (?)• Shear modulus G

• Others ?

Compression Modulus

Constantinou et al. (1992)

Compression Modulus

• For 𝐷𝑜

𝐷𝑖: 2~5, 𝐹: 0.67~0.69

•𝐻

𝐹: 0.94

• For small 𝐺/𝐾• 𝐸𝑐~6 × 0.68 × 𝐺 × 𝑆2 = 4.1𝐺𝑆2

• For 𝑆~2.0• 𝐸𝑐~16.3𝐺

Thank You