Return to index: [Subject] [Thread] [Date] [Author]

RE: Loose Bolts with DTI Washers

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Title: RE: Loose Bolts with DTI Washers

>...large, 10-year old steel structure with
>bolted connections...some of the bolts are
>falling out....either loose, missing or not tightened
>properly ... DTI washers were used and were
>installed properly, but not necessarily tensioned
>properly. The building supports overhead cranes
>and is also subjected to other dynamic loads....

>1) Other than improper installation/tightening of the
>bolts, are there any other reasons or explanations
>as to why the nuts would be loose ten years later?

Properly installed ASTM A325 and A490 bolts should not loosen or fall out. In most cases, snug-tight is OK. In the case you've described (impact, fatigue, etc.), pretensioned installation is required and is normally sufficient to prevent bolts from loosening or falling out.

>2) What is the history of DTI washers?  How reliable
>are they?  What is the likelihood that the washers
>are defective?  What tests or methods are employed
>to determine if a washer is defective?  What are the
>pitfalls of using these washers?

All four pretensioning methods (turn of nut, calibrated wrench, TC bolt and DTI) are suitable when used according to the provisions of the RCSC Specification for Structural Joints Using ASTM A325 or A490 Bolts (see it at http://www.boltcouncil.org). However, any one of these methods can and has been abused by not following those provisions.

>3) Is it possible that the bolts can be fully tensioned and
>yet the bumps on the DTI washers not be flattened as specified?

Yes, if the DTI's are significantly overstrength. Also, it's possible if they are not used in the proper place in the joint. The procedures described in the  RCSC Specification are designed to guide you away from trouble in either case.

>4) On connections that are tight, but the DTI washer
>is not properly compressed, are there tests that can
>be performed to determine if the bolt is adequately
>tensioned, after ten years?

Elongation can be measured ultrasonically (and expensively) if you're willing to remove the bolt(s). The pretension that was there can then be calibrated to that measured elongation. But other than that, I do not know of a practical or accurate means of determining pretension after the fact. Torque wrench readings are not accurate indicators of pretension, unless the torque-pretension relationship is calibrated to the fasteners being examined.

Philosophically speaking, bolt pretensioning is a bit like making a steak. It is done properly all the time by following the known guidelines. But if you walked up to a steak that's already on a hot grill and had to know when to take it off just by looking at it, your guess would be as good as mine.

For steak, the guidelines involve temperature and time. For bolts, the guidelines involve preinstallation verification of the performance of the fastener components in a Skidmore-Wilhelm bolt tension calibrator, application of the known installation procedure, and routine observation of the installations by the inspector.  Anything after the fact is either a destructive test or just a guess.

>5) What tests or methods can be employed to
>determine if the loose bolts were ever tensioned?

There may be some indication that wrenches were applied. If an impact wrench were used, it would have left peening marks on the nut. But in the end, there is no way to tell. The fact that they are fallingout, though strongly suggests that they were not pretensioned in the first place.

>6) Any thoughts on re-tightening the loose bolts.

Reuse is allowed for black ASTM A325 bolts, subject to the limitations in the RCSC Specification. Reuse is not allowed, however, for ASTM A490 bolts, nor for galvanized ASTM A325 bolts.

Hope this helps.

Charlie