Bolt shear in overhead lifting application

[Chronik]

Hi,

I am currently designing a hook that is going to be attached to a lifting beam we just purchased. The beam is rated for 2000 pounds, and we have to lift up to 600 pounds total (2 hooks, therefore 300 pounds per hook).

Attached is a drawing of the current design.

Because of the thickness of the hook, there is a gap between the bracket that attaches the hook and the bracket that is welded at the end of the beam.

I know how to calculate the shear in the 3-bolts section of the system, but how do i take into consideration the 1/8'' gap there is on both ends of the beam bracket.

The hole in the beam's bracket allows a 3/8'' screw to go thru. I was thinking of using a 3/8'' shoulder bolt as shown on the drawing, and two 1/8 (or a little less, it has to be able to swivel a bit) steel spacers to fill the gaps.

**Please note that you do not see the full beam in the drawing. Only the end plate of the beam which the hook is attached to is shown.

Thank you for your time.

 

[dvd]

I think that the old-timers referred to that as combined stress loading. Your spacers will not do much to alleviate bending moments. I would increase the thickness of the middle eye plate to reduce the bending moment. You might also look at the combined stress on the bolts to see if it is acceptable.

 

[Chronik]

Thanks for the reply,

i can't modify the center bracket, and i am aware that the spacers won't help.

i rapidly calculated using Beam 2D 69 MPa bending and 10 MPa Shear, then using Von-mises (sigma_bending^2-sigma_bending*sigma_shear+sigma_shear^2)^1/2 gives 75 MPa

do you believe this approch is conservative enough?

(used E=210 GPa for the calculation)

Thanks,

 

[badbunny]

Hi, several comments...

1) What did you assume for your bending stress? Have you assumed a fixed-fixed beam? I wouldn't because unless the bolt is an interference fit into the side plates it will provide little to no moment restraint for the very small bending deflections the bolt will be seeing. I would assume as a first pass, a pinned-pinned beam with end supports mid-way through the side plates, and a central point load. If that is no good then start working your way back to remove conservatism.

2) Remember to include a dynamic magnification factor on your lifted load unless you can demonstrate otherwise (this would involve consideration of your crane lifting torque, flexibility in the ropes, etc, to calculate an acceleration). As a starting point I would suggest you use a factor of 2, and if that is no good then try to prove it is lower.

3) Have you thought about using a standard to guide you and give some confidence in your design substantiation? BS2573 (British crane design code) is quite good (it is in process of being replaced by a new Eurocode, but hasn't been fully replaced yet, and is a lot briefer than the Eurocode!).

Hope this is some help,
Pete.

 

[badbunny]

Hi, a couple of other comments...

4) Remember to check bearing stresses on the inner beam, outer plates, and the bolt. As a quick check I usually assume half the projected areas.

5) Finally, don't forget fatigue!

Cheers,
Pete.

 

[badbunny]

Typical, I just remembered one other thing as I hit reply

6) I'm not sure what country/industry you are in, but in my experience any lifting equipment must be proof tested with recorded evidence of this. Usually we try to aim for 2x maximum design load. So in this case, with an impact factor of 2, that would be to 4 x 300lb. This may be too high for your component, in which case set a lower limit that is sensible. Could be 20% above, 50% above, whatever makes sense for proving the integrity.

Pete.

 

[ESPcomposites]

Isn't this just a typical lug/clevis problem? When evaluating such a problem, the gap often causes significant bending in the pin. The methods for this sort of problem are well documented, though do not always agree amongst the various sources. ESDU has a pretty good solution.

Some simple conservative assumptions could be used by combining pin bending and shear stresses along with pinned-pinned assumption.

To relax some conservatism, you can consider:

- plastic bending of the pin at the ultimate load.
- something "better" than pinned-pinned depending on the member stiffnesses and contact distribution (see the standard methods)
- the location of max bending may not occur at the location of max shear and therefore the combination of the max's may conservative.

Brian

http://www.espcomposites.com

 

[badbunny]

Hi, I had an idea to reduce the bending on the shoulder bolt (see attached pdf with sketch - note in the pdf I say 'pin load' but I meant 'bolt load'). Pete.

 

[badbunny]

I'd be careful about stating combined pinned-pinned bending and shear as conservative. These problems have very complex stress fields and are very sensitive to the geometry and material properties. The peak stresses are likely to be significantly higher than those predicted by the hand calc. I think Roarks has some discussion about how the hand calcs for shear become increasingly unconservative for shorter deeper beams. However, I'd say that the accepted methods of combined bending and shear are reasonable provided you can demonstrate decent elastic margin, if not you are into the territory of FE analysis. Also if you go down that route you would be slightly on your own in terms of defining stress limits, because codes/standards have not really caught up with the use of FE. Also if you do FE you will likely find some results which are hard to ratify against the elastic intent of a lot of codes/standards, ie you are likely to predict localised plasticity which the codes/standards don't allow for really. Pete.

 

[ESPcomposites]

I am sure it is different for every industry, but the aircraft industry is well aware of how to treat the problem. Since I have not heard of any failures, I assume it works pretty well.

Of the public resources, ESDU 91008 can be considered.

http://www.esdu.com/graphics/dataitem/di_91008.htm

"The consideration of the effects of pin bending in a double shear joint is also included."

Brian

http://www.espcomposites.com

 

[ESPcomposites]

Another reference is Niu's "Airframe Stress Analysis and Sizing", p.327.

Brian

http://www.espcomposites.com

 

[badbunny]

From the wording in that ESDU web page 'failure by rupture' is it coming at this from a limit load perspective? (I've not got access to the document). If so, it gets tricky trying to use this philosopy when the code/standard/industry accepted targets are based on significant margins against yield, and S-N based fatigue.

Also I don't agree with your logic that because you haven't heard of any issues that the methods we are using are working well. I think in many cases the benign plastic behaviour of materials like steel and aluminium protect us from design mistakes and lack of understanding. I think we should always recognise that our methods are mainly based on empirical observations, that we don't understand fully the underlying physics, and act with appropriate caution.

Cheers,Pete.

 

[Chronik]

Hi, thanks for all the replies.

badbunny:

1) Attached is my FBD. It is pretty much what you recommended.

2) Security factor used will be 4 so that it includes impact.

3) We have cranes built exactly like this one (products that come from Italy), they are used to lift 9000 pounds (lot bigger than what i am lifting). The gap between the center bracket and the two side brackets is 2.5" (1.25 per side approx. but this has to be done because they have to adjust the position of the crane). That being said, the pin used on those is 1.5" diameter.

4) I did check bearing stress on the top part of the outer plate as it is the most critical part of the design. The center plate is part of the certified 1 ton beam so i don't have to calculate it.

5) I do not know what i could do about fatigue in this situation. I might consider it in the 4-5 S.F. The number of cycles is not huge.

6) I am in Canada. To make sure everything is safe, i am doing the calculation 1- Manually, 2- Mathcad, 3- Beam 2D and then 4- ANSYS.

ESPComposites:

Thanks for the reply. so Von-mises is appropriate to combine both just like i mentionned in my previous post?

badbunny :

Your suggestion is exactly what i had in mind, but i am doing the calculation with the gap between the brackets to be conservative as i do not want to have to take 2 thickness and the radius/chamfer in the calculation.

Thanks for your help.

 

[badbunny]

HI, von mises is fine to combine the shear and bending, just have in the back of your mind that the average bending and shear calculations are a gross simplification of the actual stress field which is complicated, and that if you are getting close to the limit you should probably redesign or move to a more complex analysis (ie FE). With that in mind though, if you can proof test to 4x then that test and subsequent inspection should give plenty of confidence. Cheers, Pete.

 

[ESPcomposites]

Pete,

I am sure we could debate this forever, but the aircraft industry has developed many tried and true methods. The approach is to develop a method, with a physical basis, as demonstrated in the references provided. The method is then usually compared with lab test data. Once implemented onto the aircraft, it is continually monitored and inspected to make sure the behavior is as expected. If there is a failure, then it is investigated to determine where the breakdown may be. Due the nature of the safety requirements and passenger expectations (i.e. don't kill them), the process is required to be highly reliable. I can't give you an exact date, but this particular method has been around for decades. I suspect you won't find anything more comprehensive than this, in any industry.

While you are correct in saying it is not "exact", very few problems in engineering are ever exact. But then again, that is the point of the engineer...to solve the problem without knowing every inconsequential detail.

Again, being from different industries, the approaches may be different. But this is not "my" logic, it is the logic of the aircraft companies.

Brian

http://www.espcomposites.com

 

[badbunny]

Hi Brian, my comment wasn't about the logic of this method (which I use and don't have a problem with), but the optimistic/unrealistic view that because we've not seen any failures to date means there is no problem with our methods. Maybe I'm just becoming cynical with old age though! Pete.

 

[Ron]

Your condition is double shear with bending. The 1/4" gap causes the pin to have bending moments. Depending on the amount of clearance at the pin holes, it can be modeled as pinned or fixed.

Check both shear and bending, then use a unity check for the two of them.

 

[ESPcomposites]

Pete,

I understand what you are saying. There have times when I have wanted to change methods because they could be "better" - at least in theory. But the problem is then you may have to start the test cycle over again, which could take years/decades. So even if it may not be as good as it could be, is it worth the redo? Usually we just say "if ain't broke, don't fix it".

Being cynical is OK too. But if we look at failures, I think we will find that they are far more often due to misinterpretation of the method, rather than shortcomings of method itself (provided it is tried).

Sorry to get off track! I just wanted to point out that this is a relatively common problem, with known solution approaches. You can make some conservative assumptions, which are fine. If you want to relax some of the conservatisms, then there are ways to go about that, as cited early.

Brian

http://www.espcomposites.com

 

[SWmechE]

Chronik,

If you haven't already read through ASME BTH-1 and your provincial OH&S code. For example Alberta requires a 5:1 design on ultimate breaking strength.

 

[boo1]

ASME B30.20 requiring a minimum design factor of three of the yield strength of the material. ASME BTH-1–2005, Design of Below-the-Hook Lifting Devices, includes design criteria, including a lower design factor, and the necessity to address other failure modes such as fracture, shear and buckling, and design topics, such as impact and fasteners. your safety factor or 4 is within the guidelines of ASME BTH-1–2005.

There are several design elements to check, including bearing, tearout, plate strength. It appears from you model there is not much material depth in the single plate, radially from the pin. Consider the check:

3-3.3 Pinned Connections gives Static Strength of the Plates
Pt = (Fu/1.20Nd)2tbeff