Single bolt connections in steel frames 3

[umgrego2]

There has been a lot of discussion at our office over the use of single bolt connections in steel frames that we design. The two more experienced people feel quite strongly that you should not design a bolted connection that uses only one bolt. One claims that it is an erection safety issue. The other states that a single bolt connection has more prying force applied to it when the connection is in compression. Unfortunately, we cannot find any piece of code which states whether or not a single bolt connection is allowed.

Has anybody else heard about this rule? What is the background/ reasoning for not using a single bolt connection in a steel frame?

 

[TFL]

Single bolt steel connections are BAD engineering practice.
they should only be used under extremly light loads and i mean
like 500 lb. they are bad simply from a redundancy point of view. you never want to have a connection that will fail if one bolt is loose or damaged. they are right It is also an erection hazard you need one hole for a drift pin and one to hold the bolt you put in to take the drift pin out

 

[cap4000]

One bolt connections is just real bad practice. Rotation of a beam can easily occur. Use a min. of 2 at all times. Good Luck.

 

[DDS1]

This issue has surfaced several times in my career. The philosipy of my mentors and now my own has been that this is bad engineering practice. If you reference AISC manual of steel construction (at least he ASD additions) they do not include in their tables tabulated values for a single bolt connection. Also in the AISC manual (one edition at least) there is an illustration that shows a connection where there is a horizontal row of (2) fasteners in the supported member. You could rationalize that the load was to great for a single fastener so they put (2) and that the depth of section limitation forced them horizontal. Or you could rationalize that even though the load was not great and therefore the section chosen lacked depth that a minimum of (2) bolts is required and therefore forced the horizontal placement.

With all that said, I came to a personal conclusion that the supported member could rotate and therefore could be a safety issue during erection as well as rotate under years of vibrations (as we see in the industrial setting). For these reasons I have stayed away from single bolt connections. If there is printed material regarding their use I would like to be advised.

DDS1

 

[CSEllc]

Let me throw my two cents in.........
single bolt connection........BAD......
Two or more bolt connections............Good.

I like to sleep nights when I'm not working.

 

[krus1972]

Umgrego2,

You are in fact right. There is no piece of code that is known to exist that directly specifies you MUST use at least two bolts in the connections. There are not so obvious reasons in all phases of steel construction why a minimum of two bolts (2 lines of bolts in beams) should be used. The steel code book addresses steel construction in ALL phases of steel construction. If you are a design engineer you are responsible for the reasons that impact only the design of the structure and nothing more. One example is to look at the entire mechanics of a typical beam end connection.

A beam end connection is either framed into a web or a flange of another structural member. When it's framed into the flange of say a building column the connection undergoes both shear and prying forces. Shear is the obvious load the connection must resist. The prying force is the load condtion that is OFTEN mis-understood. It is best describe as a moment whose axis is half way between the center of the column and the column flange where the beam connects. The moment's value is the the beam end reaction times the distance from the column flange to the column center.

The prying force is ususally the beam-end reaction times the distance to the column center.The prying force MUST be resisted by either the end connection or the column itself.

If you decide to let the column resist the moment then the column must be designed to resist this moment at EVERY area where a beam connects into a FLANGE in ADDITION to it's other design parameters for the column application. Now your beam end connection can be a simple shear single line connection located at the beam center without no design for prying due this moment. If a beam connects to a column WEB then there is no moment and therefore there is no prying action at this type of end connection. It's just shear.

If you design building column(s) to be the structural element that resists end connection prying you can see how cumbersome it will be since some beams frame into the column FLANGE and some beams connect to the column WEB most having different end reactions producing different moments at different points along the length of your column(s). All of this just because you want to use a single line of bolts on every beam.

Thank god for the invention of the multi-line beam end connection!. If the end connection is used to resist the moment then a single line of bolts located at the beam center will NOT resist prying. The moment is resolved into both the tension and compressive forces. These forces depend on the placement of your of your bolt lines.

The TENSION is what is commonly called the prying force because it's located ABOVE the neutral access of the beam and wants to pull away from the column while. The centroid of your multi-line bolt group is placed at the CENTER of the beam with the top bolts resisting tension (prying) AND shear. The bottom portion of the bolt group resists only shear. The compressive force is applied directly to the column flange from the connection angles. Again if the beam connects directly to the web of the column there is not Prying moment or forces.

As you can see it is MUCH easier to make standard tables and standardize end connections for the BEAM END CONNECTION to resist the required prying loads. In multi story buildings it's just not pracitcal to design the columns for this.

Above is an example of why beams normally use at least two-lines of bolts but what about simple tension and compression members such as knee bracing where one bolt is sufficent from the design stand point?

Lets look at the actual erection of the element. If you take note on how most iron workers first align the holes when bolting up the connection you will notice they use a spud wrench. A spud wrench is a solid, long, and tapered rod that has an open-end wrench on the other end. They place the tapered end of the wrench into one of the bolt holes to align the connection. Once one of the other open bolt holes is close to being align they will place the bolt in it and pull the wrench out from the first hole. They will then place a bolt in the first hole and place the buts and washers on both bolts. They flip the spud wrench around and use the open end to snug type both nuts. The iron worker will move and go on to the next piece that needs to be erected and do the same. The key in erection is to keep the crane moving and reduce the time it takes for the iron worker to stabilize the connection. Placing only one bolt hole (or bolt line) makes alignment & stabilization much more difficult for the iron worker. You have to remember the crane can only ROUGH in the alignment. It's the individual iron worker who has to make the final connection alignment to place in the bolts. If you give him only one bolt he has a much harder time aligning the connection since he cannot use the mechanical advantage of a long spud wrench in the other hole to align the connection. It may be hard to imagin this but if you pay very close attention to the iron workers as they place the first bolt in a connection you will get a good feel of why at least two bolts is important fron the ERECTION stand point.


As you can see more then one bolt is not ALWAYS required by design and not always required for construction. Thats why there is no governing body that explicatly states two bolts (or bolt lines) must be used. However using more then one only makes life easier for the erectors and the designers.



Jeffrey A. Krus P.E.

 

[umgrego2]

Thanks to everyone for your thoughtful replies.

Just to let you know what my other research has provided:

The CISC responded with:

"From an erection consideration perspective, 2 bolts are the minimum required for installation purposes. Within the industry, there has always been a general rule of 2 bolt minimum for any structural connection."

However, the reply I got from the AISC referred to OSHA standards:

"You will want to check the Federal Register, as OSHA does have some rules regarding the number of bolts that must be in place before a structural member can be released from the hoisting lines (Section 1926.756). The Register can be obtained from

http://www.osha.gov.

Note that the Register does disntiguish between standard beams and columns and bracing members. You may be able to use a single bolt in bracing members."

More specifically (Direct from OSHA):

1926.756 Beams and columns.

(a) General. (1) During the final placing of solid web structural members, the load shall not be released from the hoisting line until the members are secured with at least two bolts per connection, of the same size and strength as shown in the erection drawings, drawn up wrench-tight or the equivalent as specified by the project structural engineer of record, except as specified in paragraph (b) of this section.

(2) A competent person shall determine if more than two bolts are necessary to ensure the stability of cantilevered members; if additional bolts are needed, they shall be installed.

(b) Diagonal bracing. Solid web structural members used as diagonal bracing shall be secured by at least one bolt per connection drawn up wrench-tight or the equivalent as specified by the project structural engineer of record.

So basically, OSHA is fine with a single bolt connection for diagonal bracing only. But having read what krus1972 wrote, I will be specifying two-bolt connections even for braces.

Thanks again to everyone

Mike

 

[Grouser]

umgrego2

Since it is "standard practice" to use at least two bolts, you will lose in court if you use fewer.

Also, per krus1972's accurate erection discussion, any money saved in fewer bolts and holes will be spent in longer erection time so there is no economic reason to do so.

 

[epr]

Everyone seems to be talking about beam-to-column connections. Sometimes we design frames with knee braces (or kickers) where the members are in tension or compression, or trusses where the tension and compression forces are relatively small and one bolt can easily suffice. This is not unusual, it has been done for years, and it is not subject to any significant prying action.
However, on a beam-to-column connection . . . I agree with all of you above: 2 bolts minimum.

 

[Lutfi]

My humble point of view that single bolt, even if it is adequate by calculations, is very bad practice. We are not talking about huge expense but we buy a lot of insurance. Another thing that I always do is use minimum of 3/4 inch diameter bolts in my structural frames.

This is my two cents worth.

 

[FFIELD]

Going against all the previous answers, which I agree with generally, the use of a single bolt connection is used all the time. It is called a pin connection. If the proper materials are used and the provisions of the AISC are followed it is perfectly OK (Theoretically). From a practicle standpoint I agree with all the above.
I have used pins in various structures with no adverse results. It may or may not be practicle for a given condition but since I do not know the conditions I could not give an opinion.

Do not give up on an idea just because you are told it will not work until you fully explore it. Just might be a solution for a given condition.