No announcement yet.

Bolt strength without proper torque?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Bolt strength without proper torque?

    I have a U shaped bracket that fits over a piece of wood, a glue lam beam. Another part of the assembly is then bolted over the outside of the U bracket with 3 3/4"-16 x 8" grade 8 bolts. My question is if I will be able to apply proper torque to the bolts with the wood in the middle of the sandwich? Will the wood yield over time so the bolt tension will decrease? If my torque chart is correct these should be tightened dry to 420 ft-lbs which gives a tension of 33,500 lb for each bolt. I was just curious because I would like to know if these bolt will end up in pure shear. My tables say if this is the case then the loading capacity should be decrease to 60% of what the load would be with proper torque. Anyone have any experience with anything like this?


    Hello, my name is brian and I'm a toolaholic.

  • #2
    No direct experience, but I do think that the wood will collapse a bit over time and the bolt torque will become essentially non-existent. It seems to me the best idea is to use the lower rating (60%) as the basis for design.

    You might be able to achieve the full rating at the start, and maybe it will still remain high for a week or ten years- then what, is there a structural compromise in place?
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-


    • #3
      If this is a fatigue situation, then you have reason to worry. Otherwise, you should design as if the bolts had no pre-tension, but you can assume they have their static strength, rather than their fatigue strength, safety factors notwithstanding.


      • #4
        Thanks, that is kind of what I thought. I will use the lower number just to make sure there is never a problem.


        Hello, my name is brian and I'm a toolaholic.


        • #5
          You might look at how metal parts are attached to the wood spars of airplanes for ideas on the best approach.
          Any products mentioned in my posts have been endorsed by their manufacturer.


          • #6
            You need to understand the reasons for a torque specification when tightening a bolt. The purpose of tightening a bolt to a given torque specification is to place it in tension. The purpose of this tension is to preload the joint to allow forces to act on it without separating it.

            A good example is the head on an engine. There is always a torque specification for engine head bolts or studs. Tightening to this torque stretches the bolts or studs; placing a calculated amount of force on the head to hold it down. When the engine runs, the combustion or other forces inside the cylinder produce a force that is trying to lift the head off the block. The pre-load tension MUST be greater than the force from combustion or the bolts or studs will stretch to a point where the head lifts off the block and the combustion gases or steam or other expanding gas will escape at the joint. So the engineer who designs the engine calculates the force necessary to hold the head DOWN. He adds a certain amount, most likely a percentage from experience, for a safety factor and temperature changes and other secondary reasons, and then he calculates the amount the bolts or studs need to be stretched to achieve that force. Finally, the correct amount of torque to produce that stretch is calculated and viola, you have your torque specification FOR THAT ENGINE. These two steps may be combined into one with a table or overall formula. He then selects bolts or studs that can withstand that amount of tension without failure over the lifetime of the engine. There is also a safety factor that is added when selecting the bolts or studs.

            A key factor here, is it is the joint that the torque is calculated for, NOT THE BOLT OR STUD. Selecting the bolt or stud is the last step in a logical design process.

            You are dealing with wood. Wood will compress. Wood will compress more with time if the tension is too great or there is a changing load, as in the engine. This is a bad situation for a bolt that needs to be torqued in order to work. I am not familiar with bolted wood joints, but if the joint is designed to require grade 8 bolts for sufficient strength, then my gut feeling is you probably need to redesign the joint to allow more bolts, perhaps two or four or even more times as many. It probably should be designed where a grade 3 bolt would work (gut feeling, no design expertise here). Then the torque would be a lot lower and a lot less critical.

            Look up in any church or auditorium or other large building that uses bolted wood trusses for the roof. Each joint has a lot of bolts. A LOT OF BOLTS! Each bolt is carrying only a small part of the load. Not 50% of it. Not 25% of it. They only carry about 5 or 10 % of it or perhaps even less. They are torqued less and the wood will not compress a lot with changes in the load or with age. Likewise for the lighter, per constructed trusses intended for houses and smaller buildings. They may be nailed with metal or wood plates for the joints, but there are a lot of nails in each plate. Each nail carries only a small part of the load.

            I’m sure structural engineers have rules for calculating such joints. Some research is probably in line here.
            Paul A.
            SE Texas

            Make it fit.
            You can't win and there is a penalty for trying!


            • #7
              Consider the bolts to be in shear.... and make sure to run the numbers for bearing if the wood takes any load; bolted joints in wood structures generally don't require very strong bolts, as the wood fails first.

              - Bart
              Bart Smaalders


              • #8

                Thanks for your reply. I understand that friction is what makes a bolted joint work. One of the reasons that I wanted to know was because of the fact that in any structure the holes are never perfect, so in a case like this one bolt could end up with more than its share of the load. I am not sure how to compensate for this. I was thinking that perhaps if the plates were drilled insitu then possibly it would eliminate some of the differences. I have never designed wood joints like this before so it is a new learning curve. The joints are designed so a single bolt can carry the entire shear load with no tension, but I would prefer to have the load evenly distributed. I just wanted to see if someone had experience with steel reinforced wood construction because I have none.


                Hello, my name is brian and I'm a toolaholic.


                • #9
                  Deemed - or not?

                  Has this been designed by an experienced qualified, competent, registered and insured Structural Engineer?

                  It should be.

                  In the case of the more usual common uncomplicated set-ups and loadings, manufactures have Engineers compile tables and specifications for their products, which if complied with are "deemed to comply" "as of right" otherwise a structural engineer is required to design or assess the elements and their adequacy for the loadings (to be) applied. These tables and spacifications are free.

                  Your local Building Authority and your insurance company will certainly think so.

                  "Designing" structural elements is a lot more than picking numbers and components out of the air or "books" and kicking it around on a forum is hardly the correct way to go.

                  I wish you well.


                  • #10
                    All of the critical bolted wood joints that I've seen use steel bushings/tubes in the holes so the clamping load and the shear stresses are caried by the bushings not the fastener. That includes the bolted trusses in several buildings. I've used a similar technique with plastic parts that are mounted with threaded fasteners.


                    • #11
                      You might want to check out the site for the Simpson Timber Company. They have been in the timber fastening business for years. Duffy
                      Duffy, Gatineau, Quebec


                      • #12
                        The reason bolted joints are torqued is for fatigue resistance, not lift-off, per se. By fatigue, I mean cyclic variation of stress or strain. The idea is that by putting the bolt in tension and joint in compression (equal and opposite forces per Newton), both the joint and load carry stress variations. In effect, the combined joint and bolt have much more cross-sectional area in the game, than the bolt acting alone. Because more area is in the game, variations in load give less variation in load per unit area of bolt, i.e., less stress variation. Stress *variation* is what is bad in fatigue. The bolt acts alone if the joint is loose, or if there is so much load the pre-tension is overcome such that the joint separates.

                        On the issue of hole mis-location causing one bolt to carry more load, I wouldn't worry too much if errors are reasonable. That is because the wood will crush locally until the other bolts pick up load. Localized yielding like this to achieve load sharing is expected in all sorts of engineering design situations.


                        • #13
                          Bolted wood joints in something like a truss are typically bearing connections. The capacity is derived from the bolts bearing on the wood, and the bolts bearing on the metal side plates. There can be connections with bolts are in tension, but you'd never be able to tighten the bolts to "spec" because, like was mentioned earlier, the wood will crush. The only way to "properly" torque a bolt in a wood connection is to provide a sleeve, like was mentioned, and make the bolt clamp down on the sleeve, which then bears on the wood.

                          Paul got it right, the big bolted connections you see in churches and such use lots of bolts because we want to share the loads among a bunch of small, low grade bolts that will deform a little to help load sharing. (A307, grade 2, etc..)

                          Bolt holes are drilled 1/16" oversized, and are expected to move a little bit until the connection takes a set. There are some self drilling dowels that can be used with steel knife plates (fit into kerfs in the wood) that drill through the wood, the steel, and the wood. They're very cool, German, and a little pricy for a one-off situation.

                          Metal to wood bloted connections are typically tightened to "snug". Which is best done with hand tools, just a wrench, not a big impact and not some worker with Popeye arms. Like many have said, the wood will simply crush and the pre-tension in the bolts will be gone in a short time.

                          If you just want to make sure the connection doesn't loosen up over time, consider locking nuts, cotter pins, etc...or just peen the tip of the bolt so the nut can't come off.
                          Mike P
                          1919 13" South Bend Lathe
                          1942 Bridgeport M-head Mill


                          • #14
                            Thanks for the information. The saddle over the top of the beam is actually carrying the load. The saddles were to the engineers specs. I was just wondering about the load sharing among the bolts. I am sure it isn't a problem or it wouldn't have been engineered the way it was. I was just curious because I don't think the bolts will equally share the load because the holes aren't perfect. If I take the 60% number I found in my fastener table, for a bolt in pure shear, a single bolt will still carry the entire load with a 3x safety factor. I tried to talk to the engineer again but he hasn't gotten back to me yet, maybe tomorrow. I guess it is just my nature to want to overkill on everything. I had them beef the original columns up because I didn't like the way the original design looked. Their answer was sure it is your money. I'm sure that they will beef this up for me too if I decide I don't like it.


                            ps. yes Tiff it was designed by an actual engineer and no my insurance company doesn't care, all they asked was how many square feet did you add. I have never once had them ask about engineering for any of my other buildings either. The only reason the inspector cares is because it increases the tax rolls when I add things. When they came and looked the first time the engineer had to be called to explain because the inspector was clueless. I called them on it once, when I thought the fees were exorbitant, and asked what they would do if I had a problem with a structure. Answer, nothing it is between you and your building contractor.

                            Hello, my name is brian and I'm a toolaholic.