Need a book? Engineering books recommendations...

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

Re: Bolting issues, generation II

[Subject Prev][Subject Next][Thread Prev][Thread Next]

A lot of >'s follow. It's a mix of what I (Charlie) said before, what
Majid asked subsequently, and what I've said this time.

>>Charlie wrote previously:
>>When you install a bolt, you are talking about a range of from 1/3
>>turn to 2/3 turn from snug to induce the pretension ... Rotational
>>capacity tests demonstrate that high-strength bolts can withstand
>>several turns before failure in torque-induced tension
>Majid asked subsequently:
>Is this regardless of grip length and bolt size? If we are this
>confident, where can we set a practical limit for the maximum no.
>of turns?

Essentially it is. ASTM A325 and A490 only cover diameters from 1/2 to
1-1/2 in. and the available length range only goes up to about 8 in.
(longer lengths may be available by special order, but are not very
common). So there really isn't that great a difference in the possible
range of parameters here. In any case, the most critical parameter is
for the combinations of grip and bolt length that have the fewest number
of threads remaining in the grip because this is where all the ductility
comes from. But again, we're talking about fractions of a turn being
required to induce the pretension but several full turns required to
break a bolt. With this in mind, I guess I just don't see any practical
need for a limit on the number of turns as long as the bolt doesn't
break during installation (see below for further comment on this one).

>Majid asked:
>Over-rotating can make a 10% to 30% increases in pretensioning load
>beyond 70% specified preload. I agree, this is not much of concern of
>this is only during the instalation, or under service loads where shear
>is resisted by fricion. However this additional tension will
>significantly reduces shear resistance simply to almost zero under

Pretension does not affect the shear strength of the bolt at all. It is
not at all like externally applied tension. The shank of the bolt has to
deform in order for the bolt to fail. When this deformation occurs, the
pretension is lost. An externally applied tension would still remain
however. Accordingly, shear-tension interaction is only applicable to
the latter case. This is not a guess on the part of RCSC or something
that is based upon limited testing. It is a behavior that has been
verified experimentally in every testing program that has ever been
conducted to failure with pretensioned bolts subjected to shear (or to
tension...or to combined externally applied shear and tension).
>Majid wrote:
>Also, if in tightening a bolt only an additional 1/10 of a turn
>would cause it to fail, is this bolt still accpetable. What happens to
>its margin of safety?

Are there a lot of ironworkers spending extra time turning nuts on A325
bolts five, six and seven full rotations? If they did the
pre-installation verification testing that is required in the RCSC
Specification, they would find that, depending upon bolt length and
diameter, 1/3 to 2/3 turn from snug is commonly sufficient to induce the
pretension. If they are out there cranking nuts around and around and
around, however, one of two things will happen: either the bolt will
break or it won't (duh!). If it does break, we're OK because we know the
bolt was turned too far...more importantly, they know it was turned too
far AND have to replace it. If it doesn't break, we never made it over
the hump in the plastic portion of the bolt load-deformation curve. Nor
will we ever if the joint was designed properly since the actual load
the bolt will experience in service will never exceed what it already
saw in installation. Actually, in a properly designed joint, the load
will not exceed the minimum pretension. Now the amount of stretch (prior
to failure) left in the bolt may be lower, but from a strength
perspective, it is fine for service (I don't just mean service loads

>>Charlie wrote previously: 
>>Also, when you let go of the wrench, the stress state in the bolt
>>changes from combined tension/torque to tension only, which is not
>>nearly as demanding. Add all this up and the worst stresses the bolt
>>will ever see (if designed to meet the AISC and RCSC Specifications)
>>are induced during installation.

>Majid asked subsequently: 
>Please, correct me if I am wrong. When I am turning the nut, the main
>torque is to overcome frictional torque developed between nut and the
>faying surface.

You are correct, but there is some torque on the bolt shank during
installation due to whatever friction exists on the interface of the nut
and bolt threads.
>Majid asked:
>Why shear deformations in the transfevrse direction could affect
>longitudinal deformation in a bolt? In fact even if there is the
>sligtest effect, assuming it to be due to large deformations, it must
>be in direction of increasing that tension not decreasing it. I would
>agree with you if you say AFTER removing the shear, pretension may be

Shear deformations occur on planes that are oriented 45-degrees with the
axis of the bolt. So no matter if the bolt is loaded in shear, tension
or combined tension and shear, the bolt has to elongate as it sure couldn't shrink!

>Majid asked:
>I would believe it if you give me a reason why pretensioning force
>could be all lost due to shear deformations.

I think I have explained the reason why it is so in the answers above. 

>Majid wrote:
>I found too few data supporting <Kulak's> Fig. 4.15. Either there must >be a logical reason to support that old experimental data or it was
>just matter of luck

I wouldn't claim that RCSC couldn't ever be wrong, but I can tell you
from personal experience with the individuals that make up RCSC that it
is a top-notch group of engineers and constructors who write a very
practical and useful specification. They don't recommend or endorse
anything without careful consideration and deliberation. G.L. Kulak
(main author of the Guide to Design Criteria for Bolted and Riveted
Joints) is one of the most technically knowledgable member of RCSC. So
there isn't anything I'm aware of in either the RCSC Specification or
the Guide that is based on luck.