I'm looking for a justification for the use of cotton duck bearing pads without a sole plate. I'm specifically considering their use with precast I-girders.

Say I have a two span bridge, simply supported for dead load, with expansion bearings at the abutments and fixed bearings at the piers.

The configuration for the expansion bearings that I'm familiar with is:

enter image description here

(bearing plate embedded in girder - welded to sole plate - which is attached to stainless steel sliding plate - that rests on a PTFE sheet - which is attached to the cotton duck pad)

The configuration that the Nebraska Department of Roads specifies for a "fixed" bearing is:

enter image description here

(bearing plate embedded in girder - sitting on cotton duck pad)

  1. How can this be justified as a fixed girder? Are there studies that support the reliability of counting on friction between the bearing and a steel plate?

  2. If it's not truly a fixed girder, do I have a "floating" bridge? What design considerations/concerns does this raise? If you've encountered this consideration what is your design approach?

  • 1
    $\begingroup$ I'm always amazed by the sheer amount of US terminology which means nothing to me! I've never heard of a "cotton duck pad". From a quick bit of googling I'm not really any the wiser. But relying on friction to hold a bearing together isn't unlikely, it's standard in the UK with elastomeric bearings (which may be the same as cotton duck... I'm really not sure, but probably not...?). $\endgroup$
    – AndyT
    Commented Sep 1, 2015 at 13:32
  • 1
    $\begingroup$ I think cotton duck pads are also known as 'fabric reinforced' elastomeric pads. It's thin layers of elastomer and fabric sandwiched together. But unlike steel reinforced elastomeric, they have almost no shear deformation capacity (due, I suppose, to how thin the elastomer layers are). I'd be very interested to learn more about the UK allowance to rely on friction. I think AASHTO prohibits it except for certain seismic applications. $\endgroup$
    – CableStay
    Commented Sep 1, 2015 at 13:53
  • $\begingroup$ Thanks for the additional info. Based on it having elastomeric surfaces, then its friction is likely to be enough to hold it in place. The bearing design will need a check for minimum vertical load with max longitudinal load. That said, your statement of "almost no shear deformation capacity" makes me wonder whether it is unsuitable to be used as a fixed bearing, as presumably it couldn't transfer the longitudinal shear? $\endgroup$
    – AndyT
    Commented Sep 1, 2015 at 14:00
  • $\begingroup$ @AndyT I believe the "almost no shear deformation capacity" was referring to the high longitudinal stiffness rather than being unable to handle high shear loads. It just means that if you have a forced displacement you'll get extremely high loads as with any "fixed" grider. $\endgroup$
    – Rick
    Commented Nov 18, 2015 at 16:25
  • $\begingroup$ @Rick, could be... But certainly within my lexicon "almost no capacity" <> "extremely resistant to". $\endgroup$
    – AndyT
    Commented Nov 18, 2015 at 16:28

1 Answer 1


Remember than large structures have to move or they will fracture. Think of the wear equation and the coefficient of friction in the two options. PTFE is an amazing polymer, goretex. Even if it is a dead load, there will be significant shear due to bending and environmental conditions (so rotational too). If it's welded the stress observed will be tensile and compressive with stress concentrations along the beads, less evenly distributed and more prone to fracture due to fatigue.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.