1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Bolt design for butt joint

  1. Feb 9, 2014 #1
    I am trying to select a bolt for a butt joint connection. When you guys do bolt selection, are there certain equations you use to determine if the bolt will rigidly hold the 2 parts together in the joint?
  2. jcsd
  3. Feb 10, 2014 #2


    User Avatar
    Science Advisor

    How can you use bolts for a but joint? I think we need a better description of your particular situation.

    There needs to be a plate cover over the butted plates at the joint. The bolts clamp the cover to the butted plates so friction between the cover and plates will help hold it together. That way the shank of the bolt is under tension, so not in it's weaker shear condition. For less investment in fasteners and work, that can do more holding. These days, as for the last 60 years, it is highly probable that a butt joint would be welded not bolted.

    There are manufacturers data tables on the strength of bolts and there are many older engineering books that analyse the size, number and distribution of fasteners, such as bolts or rivets, in many different situations.
  4. Feb 10, 2014 #3
    This is an image of something we'd like to do http://www.ripsdiy.co.za/graphics/woodbutjoin.jpg, except the screws you see in the pic will be bolts. In the bottom piece in the image, there will be a blind threaded hole. The top piece is just a through hole, no threads. So we want to clamp the 2 pieces together using a bolt.

    Right now, we do not want a permanent connection, so we didn't want to weld it.

    I have also attached another view of what I intend to do. Basically we're making a box. The 4 sides of the box are bolted together. The 2 holes you see are the bolt holes. Inside the box, there will a pressure of 100psi. We want the bolts to fasten the plates rigidly. I am just not sure what kind of analysis we need to do to determine if our bolt can handle this. The 100psi will induce a shear and axial load on the bolt

    Attached Files:

  5. Feb 10, 2014 #4


    User Avatar
    Science Advisor

    Google bolt strength and see what you get...

    The internal pressure of 100 psi will push the sides and the ends of the box apart. There will be a similar, if not greater force on the base of the box.

    The force on the end of the box will be the end area in square inches multiplied by the internal 100 psi. The four bolts shown will have to handle that tensile force. Each will carry one quarter of the total.

    The force on the side of the box will be the side area in square inches multiplied by the internal 100 psi. The four bolts shown will have to handle that shear force, each must carry one quarter of the total.

    Since the shear strength of a steel bolt is usually taken as 58% of the tensile strength, the box may be strongest if you put the bolts into the ends from the longer sides. But that will depend on the base and the lid attachment.
  6. Feb 10, 2014 #5
    Thanks for the reply. That is very much like what I tried. Say the side area and end area are 1 in^2. That means the shear and tensile force are 100lb. Assuming 4 bolts means each bolt should handle 1/4*100=25lb.

    Say I use a 4-40 Grade 2 (http://www.derose.net/steve/resources/engtables/bolts.html). The specs on this site say the proof load is 55ksi and the stress area is 0.006in^2. So the proof load in lbs is 330lb, which is >>> 25lb. So that means the bolt will easily handle that tensile load.
    The shear is just 0.58*330, which is still >>>25lb. Which means the bolt scan also easily handle the shear force. So a 4-40 bolt should handle the 100psi quite well. Is that the right idea?

    One thing that concerns me is the thickness of the plate with the blind threaded hole is not present in any of these calculations. Shouldn't that have some effect on how rigid the connection of this joint will be?
    Last edited: Feb 10, 2014
  7. Feb 11, 2014 #6


    User Avatar
    Science Advisor

    You can reduce the size of the bolts specified, as that will minimise machining and bolt cost.

    If steel bolts are used in steel then once you have screwed them in two bolt diameters you should be secure. There is absolutely no advantage in going beyond 4 bolt diameters.

    But, if the box is made from a material that yields at a lower point than the bolts, you will need to change the design to use a bigger bolt so as not to pull the thread out of the box. Knowing the ratio of yield strength of box material to the yield strength of bolt material, you can select an oversize bolt that will hold in the box material. You might also consider using a deeper threaded hole.

    The bolts only hold the corners together. If box wall thickness is too thin it will buckle. There are guidelines as to how far a bolt must be from an edge. They are usually prescribed in the standards.
  8. Feb 15, 2014 #7
    This can be answered both in theory and practice.

    It might appear in theory there would be no force transferred through the bolt. Instead the axial force is entirely transferred through the end of the two butted plates. Yes, but the plates are prone to buckling. So the cover plate can be sized to resist the buckling of the butted plates. Thus, the bolts can be sized to transfer the buckling force to the cover plate.

    In practice, the structural steel code will give direction on the minimum quantity of bolts required.

    A classic example of bolted butt connection is the splicing of building columns.
  9. Feb 15, 2014 #8


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Ouch. Be careful what you are using for areas in your calculations. It is the surface area of the end and side pieces that determines the load on the bolts, not the end cross section areas. If your box is say square with 10 inch long side and end pieces each 5 inches high, your area to use is 50 in^2 not 1 in^2.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook