Atypical Bolting Application Qestions

[jdonville]

Note:
**I have already posted this in the Mechanical Engineering Other Topics and Structural Engineering Other Topics fora inadvertently. I will direct replies from those fora to this thread.**

I am working on connecting a planar frame to the counterweight of a heavy crane. The current scheme has a single line of high-strength bolts connecting through both of the cantilevered frame members up into a tapped plate (call them the "Bolt Plate"s) that will in turn be welded to the bottom of the counterweight.

The crane manufacturer suggested 1-inch diameter Grade 8 bolts. From what I can determine, this closely corresponds to ASTM A354 Gr BD, which I understand is equivalent to the material requirements for ASTM A490 bolts.

Owing to the required moment capacity at the point of attachment, the frame member is rather deep (9 inches). So, to clarify, the bolt will pass through the 9-inch deep member and connect to the Bolt Plate. Based on the plate and bolt materials, the required length of engagement is about 2 inches. Throw in an allowance for washers and I'm looking at a bolt length on the order of 12 inches overall, with a thread length of at least 2 inches.

I have chosen a bolted connection over welding to permit easy removal of the frame should it ever need to be done for using the crane in a different situation.

I anticipate that the bolts will be loaded primarily in tension, and will be subject to cyclic "impact" loads as the load on the frame bounces while the crane travels, rotates or hoists loads.

Shear loads should be infrequent and of much lower magnitude and duration than the tension loads.

**Questions**

How should I analyze the capacity of the bolts?
Is pretensioning required?
How will the pretension affect the capacity of the bolt?
Can a bolt this long be pretensioned using the bolt head?
How should I incorporate fatigue into the analysis?

Most of the literature that I have relates to typical structural bolted connections using nuts, and may not be applicable. Any guidance and assistance is appreciated. I have access to the 2005 AISC codes and materials and AASHTO 2002 (HB-17) if you would like to quote code references.

Jeff

 

[desertfox]

Hi jdonville

Firstly yes analyse the bolts and the tapped holes for correct thread engagement length as the plate material may well be weaker in tension that the bolt material.
You need to analyse the bolts for all the loads that they will see in service combined or otherwise and including fatigue. These sites may help:-

http://www.mech.uwa.edu.au/DANotes/threads/home.html

http://www.utm.edu/departments/engin/lemaster/Machine Design/Lecture 29.pdf

http://www.roymech.co.uk/Useful_Tables/Screws/Preloading.html

regards desertfox


These topics can be found at:-

http://www.utm.edu/departments/engin/lemaster/machine_design.htm

Mechanical Fasteners - Tensile and Shear Stress Areas Notes Answers
29 Mechanical Fasteners - Tension Connections Notes Answers
30 Mechanical Fasteners - Torque Vs Preload Relationship Notes Answers
31 Mechanical Fasteners - Shear Connections Notes Answers
32 Mechanical Fasteners - Combined Tension and Shear Loading Notes An

 

[jdonville]

desertfox,

Thanks for the references.

Do you know of any that are more specific to the design of a screw connection (no nut, per se)?

Jeff

 

[desertfox]

Hi jdonville

From the earlier links read through this one:-

http://www.utm.edu/departments/engin/lemaster/Machine Design/Lecture 28.pdf

It goes into internal and external threads of different materials and length of thread engagement, which is applicable to a tapped hole in a plate.
The machinery's handbook as thread engagement lengths in but there is an error in the formula in my book

regards

desertfox

 

[MintJulep]

I'm having difficult seeing your connection in my head from your description.

A picture would be nice.

 

[jdonville]

Folks,

Here is a drawing. An end-on view of the beam is at left.

Jeff

 

[MintJulep]

I don't particularly care for the design.

Preloading the bolts will tend to crush the tube. (Is there a seam in the tube?)

Most of the load will be taken by the first bolt, and will have a further tendency to crush the tube. High stress intensity local to the first bolt - seems like the failure mode is likely to be buckling, sudden and catastrophic.

If you don't get a bucking failure then local crushing of the tube will cause total loss of preload in the first bolt.

I would much prefer welding an angle to the bottom of the weight, and a plate on the side. Tube in between them - bolts in double shear.

 

[desertfox]

Hi jdonville

As already pointed out not the best design tube will have a tendency to crush in the area of bolt or screw head.
Are the only bolts holding the frame together those shown at the bottom of your picture.
How is the frame/counter weights connected to the crane?


regards

desertfox

 

[MintJulep]

Thinking more about this, the weights are probably cast from the nastiest cheapest whatever that could be found.

I'm not sure I'd trust anything welded to them.

 

[jdonville]

The counterweights sit on a 4-inch thick 36ksi plate located at the bottom of the counterweight stack. Pistons attached to this plate lift the counterweight stack into the working position.

The HSS beam tube (46ksi) will be fitted with appropriately-sized pipe stiffeners at the location of the bolts. Plate washers will spread the load from the boltheads over the stiffeners and web of the HSS. All this is shown on the drawing, althought the pipe stiffeners are not explicitly called out.

Due to openings in the bottom counterwight plate, which only covers a portion of the bottom counterweight, there is limited room to attach fittings. I cannot show the plate dimensions due to this information being considered confidential by the crane manufacturer.

Jeff

 

[Tmoose]

"Is pretensioning required?
How will the pretension affect the capacity of the bolt?
Can a bolt this long be pretensioned using the bolt head?
How should I incorporate fatigue into the analysis?"

Pretensioning a stretchy bolt across a stiff joint can reduce the stress variation in the bolt to less than 10%, making fatigue just about a non-issue. That's a Pretty good deal for free, especially since doing it the "other" way means >trying< to quantify the uncertainties about thread form and finish will eat up a batch of time and require many essentially meaningless reports be created and presented.

I've been living in close proximity to several MIL specs for the last few years, and now I'm afraid to use bolts with UTS > 120,000 psi outdoors.

I share other's concerns about making a good bolted joint across/through a hollow rectangular beam. The actual (welded?) tube stiffener could move your fatigue issues to a pulsing ring in the tube wall around each stiffener.

 

[jdonville]

Tmoose,

My problem is that I don't know what design code(s) or guidance to use for the bolted joint, as AISC usually doesn't consider joints with large grips or joints that don't use a nut, and the mechanical engineering literature that I have been able to peruse - surface deep only, I'll guess - doesn't explicitly reference ANY formal design code that I can tell.

I can do the static structural analyses fine - that's all AISC, but the mechanical or machine design stuff seems a "Black Art" to me.

Some guidance - even MIL specs - would be helpful to me right now.

Jeff

 

[jdonville]

MintJulep,

Could you please elaborate on your comments?

The bolting scheme comes as a sugestion from the crane manufacturer. I am looking for guidance and maybe some help in analyzing the scheme for myself so that I am comfortable with the results so I can confirm or refute the manufacturers analyses prior to going to the expense of fabrication and testing of a full-scale version.

Jeff

 

[desertfox]

Hi jdonville

The only difference in the nut or tapped hole is:- one :-material grade and two:- the thread engagement.
You can analyse thread stripping, or bolt failure based on the loads the bolts and tapped holes see in service just as you would for a nut and bolt.
Fatigue can also be analysed provided you know the service loads and there fluctuation, thats why I provided the reference sites.
From your diagram I can't tell how those bolts are loaded because it merely shows the counter weights sat on the beam
with bolts going through it into the tapped plate which is rigidly connected to the counter weights, there's no indication how the frame is fastened to the crane, however if the bolts see tension from the counter weights bouncing
then yes you have a cyclic load but what valure of load?
If the bolts stop the counter weights falling over due to rocking then the outer bolts will see the most load, but without more information I don't think we can help much more.

regards

desertfox

 

[MintJulep]

Well, let's start with:

The joint is statically indeterminate. Consequence of three bolts.

So, let's imagine the same configuration, only with two bolts. If we assume that the bolt nearest the load can't really contribute very much moment, that front bolt carries ALL of the vertical load.

With your cantilevered load, there obviously MUST be a net downward force between the first bolt and the crane to counteract the moment that the load produces.

So clearly the front bolt caries most of the vertical load.

So, if we conservatively estimate that it carries ALL the vertical load, you can construct an appropriate shear and bending diagram, calculate the stress in the member at the bolt location, and size the washer appropriately - maybe.

A pipe stiffener helps the concern of crushing the tube, but introduces variables related to the HAZ of the (assumed) weld that connects the stiffener to the member.

The counter weights' design criteria is "be heavy". I am certain that nobody gave any consideration to material properties of the end result when throwing scrap into the crucible to cast them. There is simply no way of knowing what load a weld to this stuff may support.

 

[jdonville]

Mint,

As I stated before, the CW stack is entirely supported from beneath by a fy ~= 36ksi 4-inch thick plate roughly equivalent to ASTM A36. This is the plate that we would be welding to, so I have no concerns about the quality of the plate for welding purposes.

desertfox,

Essentially, the frame is hanging from the bottom of the counterweight support plate as described above. I'm not crazy about it, either, but this is from the crane manufacturer.

Thanks to you both for bearing with me.

Jeff

 

[desertfox]

Hi jdonville

If the frame is not connected anywhere else then the bolts
only see the mass of the frame, because if the weights jump
then the whole frame moves with it and I find that difficult to imagine particularly coming from a crane manufacturer.
Who is responsible for the frame design as like you I have serious doubts about it.

desertfox

 

[jdonville]

desertfox,

There will be at least 35000 lb of counterweight excluding the frame and the load on the frame, so if the counterweights jump, we have all new problems.

I am responsible for the frame design (it will be stout). It's the connection to the crane that I am most concerned about at this time.

Jeff

 

[MintJulep]

http://www.youtube.com/watch?v=cxM83_hHJC8

 

[desertfox]

Hi jdonville

Okay the counterweights don't bounce :->
So the c/w support plate is already fastened to the crane?You are merely connecting a fabricated frame off the support plate with the bolts you describe and therefore your bolts will only see the mass of your fabricated frame
and any external load which is applied to that frame.
You have not provided any information regarding :-

1/ what does your frame actually do?

2/ the mass and the centre of gravity of your frame as
designed ?

3/ What external loads does your frame actually see ?

4/ Physical dimensions of your frame

Is there no other additional place you can fasten the frame
to the crane?

Its difficult without further information as listed above, to tell you what loads those bolts will see but I suspect they won't all see the same load, as there will be offset tensile loads generated by the frame mass and center of gravity by its own geometry, inaddition fatigue loads by
cyclic loading due to crane movement(guessing)or external
loads on frame mentioned earlier.
Post so drawings of your frame and we might be able to help further.