Anchor Bolt Proof Load

[dbags]

We have specified the use of ASTM A449 2.5" diameter galvanized anchor bolts for bearings supporting a box girder. We specified that the bolts shall be tightened to a proof load of 449KN. My question is two fold:
First, the contractor has complained that this load is astronomical and if it could be reduced. Is this value of proof load unreasonable?

Secondly, the contractor needs the proof load in terms of a torque value and insists that there is no means of converting this proof load into a torque value. Is this true?

Thanks in advance.

 

[rowe]

From the little research I've done, the proof load for A449 bolts (> 1 1/2" dia) is 55ksi. For a 2.5" dia. bolt, approx. area 4.9 sq.in. times 55 ksi = 270 kips x 4.448Kn/kip = 1200 Kn.

A proof load is required for a slip-critical connection - the pretensioning assures the presence of a normal force to develop the friction force. If your anchor design doesn't require a slip-critical resistance, perhaps you can specify the anchor bolt be snug + 1/2 turn - then burr the threads to prevent the nut from backing up. If you need lateral resistance, consider a bearing connection if the DOT allows.

Equating torque to bolt tension is possible (for those who - unlike me - haven't forgotten all their calculus and dynamics lessons) but it also is dependent on the friction coefficient. A direct tension indicator (DTI) is probably the best method of determining the bolt tension in the field, but I don't know if they are readily available for a 2.5" bolt. You might contact some of the bolt manufacturers to determine what's available or typical for tensioning a 2.5" dia bolt.

good luck.

 

[Qshake]

Slip critical seems a bit much for holding a sole plate down onto the bearing material. Anchor bolts are normally designed for bearing only and some moment if the bolt is extended to an upper sole plate.

Anyway, on to the question of relating tension to torque. To do this you have to have a Skidmore-Wilhelm and a torque wrench. You will be able to calibrate the torque wrench for your application.

One caveat about the torque wrench is that most are of the socket type and I don't know if the socket will fit over the bolt to fit around the nut. Especially if you have much bolt sticking up past the nut. Another concern is the SW and whether it will accomodate the large diameter bolt and one without a head.

 

[VoyageofDiscovery]

Hi dbags,

I have never heard of proof loading anchor bolts for bearings. Typically, I design the substructure for uplift and spec the required anchor bolt.

Regards

VOD

 

[structuresguy]

Well, another option if you get really stuck would involve acquiring a donut style load cell with high enough capacity and large enough hole size for you needs. You could then place the load cell between your bearing and the nut, mark the nut location at snug tight, then wrench it down until the load cell measures your required pre-tension. Now mark the nut location, and figure out the number of turns from snug tight required to achieve the required tension. Back off the nut, remove load cell, replace nut, make snug, now turn the calculated numbers of turns.

This would be maybe a bit of a rough way of doing things, but would probably work well enough if nothing else does. You may want to load/unload the same rod a few times to see if creep, relaxation or polishing of threads changes the required number of turns.

Also, good idea would be to make sure all threads are clean and well lubricated with a high pressure lubricant prior to tightening nut. This way you minimize any polishing of thread effect on the # of turns, and reduce the torque required to turn the nut.

The beauty of doing it this way, is you never have to calculate some arbitrary torque value, nor do you have to find and calibrate a torque wrench big enough for 2.5" rod.

Finding a suitable load cell shouldn't be too difficult. Universities and test labs all use load cells, and probably would be willing to rent one out, or provide tech support to do the tests themselves.